Venture capital tends to cluster around visible pain. Founders build for the problems they know, investors pattern-match around the markets they recognize, and entire categories of work remain untouched simply because they sit too far outside the software imagination of San Francisco and New York.

Eric Theis thinks that is precisely where some of the best AI opportunities now live.

At Bling Capital, the seed firm known for backing companies at the earliest stages and helping them scale product-market fit, Eric has been drawn to a class of businesses that rarely make it into the glossy narrative of startup innovation: third-party service providers, outsourced operators, and managed-service firms handling complicated but non-core work for the rest of the economy. These are not glamorous markets. They are the oddball corners. International subsidiary compliance. Down-market credit servicing. IT management. Revenue cycle operations. Patient access. The kinds of workflows that are essential, expensive, exception-heavy, and historically too messy to fully automate.

For years, software could handle maybe 95 percent of the task. The last 5 percent, the strange edge cases, the one-off exceptions, the judgment calls, kept humans firmly in the loop. That last layer mattered more than it seemed. It is what turned software into a tool rather than a replacement. It is what created armies of people managing exceptions behind the scenes. And it is what left billions of dollars of labor-heavy infrastructure largely intact.

Eric’s view is that AI changes that equation. Not everywhere, and not cleanly, but enough to make a new category of company viable: startups that do not merely sell software into services businesses, but increasingly sell a better outcome itself.

TL;DR

• The core thesis: AI is opening a path to automate neglected, exception-heavy workflows that have historically lived inside BPOs, MSPs, and service-heavy infrastructure businesses.

• The real unlock: Previous software could not reliably handle the last layer of weird edge cases. AI can increasingly manage that complexity from the start.

• Where to look: Markets that are critical but unglamorous, labor-intensive, geographically or culturally distant from tech hubs, and plagued by inconsistent delivery.

• What winning looks like: The best companies may not sell seats or tools. They may sell the completed task, deliver higher more consistent quality, and capture higher take rates as a result.

• What could go wrong: If enterprises build more in-house than expected, if margins collapse through commoditization, or if trust-heavy human relationships remain resistant to automation, the thesis weakens.

The Last 5 Percent Is Where the Opportunity Lives

One of the most useful ideas in Eric’s framing is that the barrier to automation was never the obvious part of the workflow. It was the strange part.

His perspective is shaped in part by time at Flexport, where he saw how fast-growing businesses can become trapped by operational complexity. The software could handle a great deal, but not everything. The result was a growing human layer built to catch exceptions and keep things moving. Over time, that creates what he describes as a kind of spaghetti process: manual workarounds, bespoke fixes, and operational logic scattered across people rather than encoded cleanly into the product.

That experience matters because it leads to a sharper definition of what AI may actually change. The promise is not simply that models are more intelligent. It is that founders can now attempt an end-to-end MVP in categories where the edge cases used to make that impossible or at least deeply impractical.

In previous generations of software, companies often started by filling the gaps with humans and hoped they would automate later. Many never caught up. The business scaled faster than the system. AI introduces a different possibility: start with a product that can absorb more of the weirdness from day one, then add humans only where absolutely necessary rather than building the company around them.

That is “why now” in its clearest form. The technology is getting good enough to handle more of the exceptions that used to break the model.

The Most Interesting Markets Are Often the Ones Tech Barely Sees

Eric keeps returning to a particular type of market. These are not categories founded by people solving their own consumer pain point. They are markets that sit outside the lived experience of most venture-backed founders.

He describes them as “neglected workflows in parts of the market that are not close to San Francisco or New York.” That line is more than geographic. It points to a deeper bias in startup formation itself. Founders tend to attack problems they personally encounter. That leaves a long tail of operationally important work untouched, especially in industries that were offshored, fragmented, or long ago dismissed as boring service categories.

Many of these workflows already have providers. That is the point. There are BPOs, MSPs, accounting firms, call centers, implementation partners, and specialty operators handling them today. But the persistence of those providers often signals less that the problem is solved and more that software has not yet been good enough to displace the labor. Eric is drawn to workflows that are essential but not central to the customer’s identity, categories where human-heavy service providers have accumulated complexity and mixed-quality delivery rather than product love, where incumbents have little economic incentive to invest in R&D, and where the market is large enough to matter even if it never looked like a classic venture category. These are not tiny curiosities. They are often deeply embedded pieces of economic infrastructure where success depends on “which person you happen to get”.

The Best AI Businesses Here Sell Better Outcomes, Not Tools

A central question in this category is whether startups should sell software into the existing service layer or replace that service layer outright.

Eric thinks both approaches can work, but his preference is clear. The stronger business is often the one that owns more of the workflow and delivers complete high-quality outcomes.

That distinction sounds semantic, but it changes the economics of the company.

A tool that helps an accountant move faster still leaves the accountant at the center of the transaction. The output quality depends highly on which accountant uses the tool. A system that delivers the finished compliance work begins to look less like SaaS and more like labor automation. It can justify higher pricing because it is not just improving productivity. It is taking responsibility for the task. And doing a consistently better job.

This is part of a broader shift that shows up repeatedly in AI investing right now. The value is moving away from access to software and toward completed work. Not software as a seat. Software as consistent execution.

Eric sees that dynamic across categories. In credit servicing, the appeal is not merely better yields, but an AI-first system that is consistently compliant without humans by default. In the healthcare revenue cycle, the prize is not just better visibility but the ability to handle the intelligence and actions required to navigate a deeply adversarial administrative process. In IT service management or international tax infrastructure, the opportunity is not another dashboard. It is a more vertically integrated system that can absorb the operational burden of managing a low-quality outsourced provider.

The company captures more value because it removes the volatility of labor.

Why the Incumbents May Not Be the Real Threat

One reason this thesis is compelling is that many incumbents are not simply slow-moving software companies. They are service organizations that neglected these workflows in the first place.

That creates a different kind of opening.

If a category is too small, too messy, too down-market, or too operationally annoying for the incumbent to prioritize, then the startup is not competing against a well-funded roadmap. It is competing against indifference. That is a much better setup.

Eric is explicit that some incumbents will attempt automation, especially in more upmarket segments. But in many of the markets he is drawn to, the economics have already signaled that this work is not strategically important enough to merit serious internal investment. If it was not worth staffing properly, it probably was not worth rebuilding from scratch either.

This is especially true in the more culturally distant corners of the economy, the places where startup founders and coastal investors have the least native exposure. The opportunities may be discoverable from the coasts, but they are often not legible there at first glance.

Healthcare as the Extreme Version of the Thesis

Healthcare appears in the conversation as both an example and an escalation.

Eric does not present a single sweeping thesis for all of healthcare. Instead, he points to a system full of data complexity, regulation, edge cases, adversarial incentives, and administrative labor. That makes it fertile ground for the same AI logic, though with higher barriers and slower adoption cycles.

Bling has already made a revenue cycle management bet in Adonis. The significance there is not just that the revenue cycle is large. It is that it sits inside a system where enormous amounts of value are trapped in administrative combat between providers and payers. If AI can manage more of the intelligence, action-taking, and exception handling in that process, then the company begins to look meaningfully more powerful than a conventional software layer.

He also points to adjacent categories like patient access and hub operations in pharma, areas where huge budgets are already spent helping people navigate systems that were made operationally difficult on purpose. In those markets, AI may not remove complexity so much as become the new weapon for dealing with it.

What makes healthcare especially interesting is that it shows both the potential and the limits of the thesis. In some areas, like enterprise health systems, the barriers to entry are enormous. Procurement is slow, compliance is intense, and new AI vendors face committee-heavy buying processes. In other areas, like dental, there may be room to rethink the model itself.

His example of Wally is telling. Dental practices are often organized around the economics of surgery and equipment payback. Wally’s bet is that software and a different operating model can simplify care around prevention and non-surgical treatments. It is not AI in the science-fiction sense. It is AI and software making a different business design possible.

That is an important nuance. Sometimes the opportunity is not replacing labor in the existing system. It is building a cleaner system because software can finally support it.

Where Durable Advantage May Come From Now

One of the more interesting parts of the conversation comes late, when Eric acknowledges there are real unknowns around defensibility in the AI era.

Historical durable advantages may matter less than they used to. Integrations, once a genuine advantage, are easier to build. Some marketplace dynamics may weaken as agents make it easier to navigate around aggregators. Scale alone may not defend margins the way it once did.

That uncertainty matters because these businesses will need staying power somewhere.

Eric’s answer, or at least his directional instinct, is that the real advantage increasingly lives in the labor automation itself. If a company finds a network effect in owning a workflow for many related businesses, is the “just use it” trusted provider for a full category, and perhaps owns the payment rails or operational system of record, it becomes more than a software vendor. It abstracts away the pain entirely, and the customers stop thinking about replacements.

That is more durable than yet another feature set. But it is still a live question. He is unusually open about the fact that some of these assumptions may prove wrong.

Where the Thesis Could Break

Eric does not describe this as inevitable. He gives at least three reasons it may not play out as expected.

The first is enterprise behavior. If large organizations lean further toward building in-house rather than buying specialized solutions, then many of these startups may struggle to become core vendors.

The second is margin compression. AI may make certain forms of service work more efficient without creating enough differentiation for startups to capture durable economics.

The third is relational. Some service businesses are not just operationally complex. They are human-trust businesses. If the customer experience still depends on a person, or if buyers simply do not trust autonomous systems in emotionally or financially sensitive moments, then the market may resist full automation longer than expected.

Those caveats strengthen the thesis rather than weaken it. They make clear that this is not a generic AI-for-services enthusiasm. It is a much more specific bet about which kinds of complexity are now tractable and which kinds of trust can actually be transferred to software.

What Eric Is Really Betting On

At the deepest level, this is not a thesis about software efficiency. It is a thesis about where the market has quietly tolerated labor because software previously failed.

That is why the oddball framing works.

These markets were not untouched because they were too small. Many are large. They were not untouched because they were unimportant. They are often mission-critical. They were untouched because they were too strange, too fragmented, had too-low margins, too low-status, too geographically distant, or too exception-heavy to fit the standard startup script.

AI makes those categories newly interesting not because it makes them sexy, but because it makes them buildable.

And for investors willing to spend time in the parts of the economy the venture market usually ignores, that may be where a meaningful share of the next wave of alpha lives.

More from Uncovered Media

As a solo GP managing over $1 billion, Elad Gil operates with a density of hits (including Stripe, Airbnb, Coinbase, and Notion) that most multi-partner firms would envy. In his recent conversations on the No Priors Podcast, Elad, along with his podcast co-host Sarah Guo, challenges the traditional wisdom that heavily regulated industries are technological laggards. Instead, he argues that medicine and law are the actual vanguard of the AI revolution, driven by a desperate need to scale expert reasoning.

TL;DR

Professional Fields are the Real Early Adopters of AI: Contrary to historical trends, law and medicine are rapidly adopting AI because they are text-heavy and reasoning-constrained.

Software Becoming as Ephemeral as Media: The era of durable SaaS is being replaced by software that behaves like media; it is cheaply generated, transient, and personalized on the fly through “vibe coding.”

The Reasoning Revolution: Reasoning models are transforming every industry from biology to robotics by moving beyond simple text-in, text-out to complex logic trees.

Organizational Over Individual Productivity: The next frontier isn’t just making a lawyer faster; it’s building collaborative AI systems that transform how entire firms operate at scale.

Elad Gil

Elad operates one of the largest solo GP venture funds ever built, raising over $1 billion while maintaining a single-partner structure. It’s also reported that he’s actively raising another fund with a $3B target. He has an unusually dense hit rate across breakout companies like Airbnb, Stripe, Coinbase, Figma, and Notion. He treats his content creation as a key strategic advantage; he turns his public insights into a roadmap that founders use to build their own companies, creating a scalable portfolio-support platform.

Elad’s model is built on high conviction and the ability to move with the speed and agility required to navigate the volatile AI landscape. He prioritizes founders who are focused on foundational shifts in capability: those who look past immediate market limitations to build toward the future of AGI. By operating alone, he avoids the friction of large committees; this allows him to learn and iterate alongside the technology.

His focus on cognitive infrastructure aligns with his background as a researcher and operator. Whether it’s investing in foundation labs like Mistral or vertical applications like Harvey, Elad is betting on a fundamental shift in software business models. The transition isn’t merely about faster tools; it’s about a total transition from selling software access to selling autonomous labor. He argues that we are moving away from selling “seats” to selling “units of cognition.” Consequently, enterprises will increasingly pay for the actual outcome or the completion of a task rather than just the tool used to perform it.

Individual Thinking & Background

Elad Gil’s perspective is uniquely shaped by a career spent at the intersection of deep research and hyper-scale operations. With a PhD from MIT in biology and degrees in math, he has seen “existence proofs” of radical change in biology that he now applies to machine learning.

His journey through Google and Twitter gave him a front-row seat to how platforms capture value during technology shifts. He was working on AI and mobile at Google long before they were mainstream; furthermore, he ran AI-centric product organizations at Twitter. This dual lens, the scientist’s rigor and the operator’s tactical sense, informed his “High Growth Handbook,” which founders now use as a playbook for scaling. Elad operates on a core scaling principle: that once a breakthrough occurs, growth becomes a function of scale rather than reinventing the wheel. His conviction in AI is based on a decade of observing scaling laws and the intuition that “anytime you make that initial breakthrough, you can usually just scale and things keep working.”

Deep Dive into Investment Theses

These insights are drawn from Elad’s conversations on No Priors with co-host Sarah Guo and guest Gabe Pereyra, co-founder of Harvey.

Thesis 1: The Professional Field Acceleration

The Opportunity: Traditional venture wisdom holds that law, medicine, and government are “laggards” that take a decade to adopt new tech. Elad’s thesis is that AI has inverted this. Because these fields are text-heavy and reasoning-constrained, they are actually the early adopters of the current AI wave.

The Rationale: In these industries, the primary input is expert time. AI doesn’t just offer marginal efficiency; it allows for an immediate increase in the “volume of expertise.” Elad points to the rapid deployment of systems like “GenAI.mil,” which reached 1 million users in 30 days; this is a speed unheard of for government software. The technology has finally evolved to meet the specific, complex logic required for high-stakes professional work.

The Conversation: The debate centers on the “future of the firm.” If AI augments or replaces the work of junior associates, how do law or consulting firms train the next generation of partners? Elad notes that while AI can explain a merger agreement’s structure, the high-level strategy and client judgment remain human domains. He acknowledges the risk; “shrinking the bench raises questions” about where the next decade of experts will come from.

The Winning Profile: Elad looks for companies that move from individual productivity tools to “organizational productivity” platforms. The winner isn’t a “wrapper” around an LLM; it’s a system that hooks into billing, governance, and data rooms to power the team’s end-to-end operations.

Thesis 2: Software Becoming as Ephemeral as Media (The Vibe Coding Shift)

The Opportunity: Elad argues we are moving away from an era where software is a durable good, something you build once and rent via SaaS, to an era where software is becoming as ephemeral as media; it is something more like a “content stream” that is rendered on the fly.

The Rationale: This is driven by “vibe coding,” where product managers skip the months-long requirements and mock-up process. Instead, they “vibe” a functional UI into existence in an afternoon. If an AI can generate custom logic for a specific task in seconds, the software begins to act more like a content stream than a static tool; consequently, the traditional “moat” of SaaS (the code itself) evaporates.

The Conversation: This shift threatens the “pure-play” software revenue model. If anyone can generate a tool for a specific task on the fly, why pay a monthly subscription for it? Elad posits that while “products” (simple tools) are replaceable, only “platforms” (where the data and collaborative tissue live) remain durable.

The Winning Profile: He looks for applications that focus on “proactive integration”: systems that extract user intent and utilize the full context of a firm’s data to complete tasks before the user even asks. The goal is to move beyond the chat box and into “proactive AI” that knows what the organization knows and acts on it.

Conviction and Skepticism

Areas of Skepticism: Elad sees two major red flags for AI tools:

• Just repackaging AI: He avoids companies that merely reformat data from a language model. If the base AI gets smarter and replaces your product, you have no competitive moat.

• Changing too much at once: He warns against forcing a massive redesign of how people work. The best tools fit seamlessly into existing processes rather than trying to blow them up entirely.

Risk and Regulation: Despite being a “future-forward optimist,” Elad is vocal about the risks of drone-based warfare and the politicization of AI. He expects AI to become a major point of discussion in upcoming elections, with public sentiment potentially turning against the technology as it becomes more transformative.

Looking Ahead

Elad is looking beyond language models to “foundation models for everything”: physics, biology, materials science, and simulation. He predicts that “reasoning systems” will soon revolutionize industries like drug discovery by designing full-length antibodies zero-shot on a computer.

The Final Word: “AI today is like a really eager intern—sometimes it knocks it out of the park, and sometimes the work is bad. Over time, it’ll get better.” (conversation with Lareina Yee at McKinsey)

More from Uncovered Media

The cybersecurity landscape has reached what Jake Storm describes as “the worst of both worlds”. For an attacker, the cost to launch a strike is at an all-time low. Conversely, for the defender, the cost to protect an organization is only increasing.

This asymmetry is driven by our current transition into the AI era. While the cloud wave focused on a new storage layer, AI introduces non-deterministic agents; these are systems that act on a user’s behalf in ways that aren’t always predictable. Jake views this not just as a new challenge but as a total refresh of the cybersecurity stack. As he sees it, we are currently in the innovation age of AI, a period of exponential vulnerability growth that requires a fundamental shift toward proactive, automated prevention.

TL;DR

The Buyer Shift: Security spending is moving from IT departments to highly technical engineering teams who favor specialized depth over broad, shallower platforms.

Innovation vs. Optimization: We are in the innovation phase of AI where threats grow exponentially. This contrasts with the current optimization phase of the Cloud, where threats grow more linearly.

The Shift Left: The next generation of winners will move security earlier in the development process, preventing vulnerabilities during the code generation phase rather than scanning for errors after deployment.

Diminishing Services: AI is beginning to absorb the traditional cybersecurity services market.

About Felicis

Felicis is a venture capital firm highly focused on the early stage: 94% of the current portfolio investments are made at seed or series A. The firm leads or co-leads most of their investments (83%), and is among the top 8 VC firms with a track record of backing 8 mega-unicorns ($5B+ exits) at Seed, A, or B in the last decade. The firm’s track record includes early bets on industry-leading unicorns such as Adyen and Plaid, alongside high-growth innovators like Tines, Semgrep, Supabase, ConductorOne, Terra Security, and Observo.ai.

A core pillar of the Felicis strategy is investing in reinvention. They look for companies that don’t just add a feature but rethink an entire category for a new technological paradigm. In cybersecurity, this translates to a belief in a built-in refresh cycle that occurs every five to nine years.

Jake’s focus on cybersecurity is a direct execution of this refresh cycle thesis. He views the current AI wave as a more significant catalyst for security than the shift to the cloud was, making this the time to find the next decade’s foundational security companies.

About Jake Storm

Jake’s perspective is rooted in a childhood spent moving between various regions of the U.S. and Brazil. This early exposure to diverse cultures taught him that trust is built on finding commonalities, a trait that informs his relationship-driven approach to venture. Before joining Felicis, Jake focused his lens on enterprise software and security at IVP, Jefferies, Qualtrics, and Zuora.

For Jake, cybersecurity is a mission-critical function that offers the feeling of us against the world. He is drawn to the sector because it is one of the few areas where budgets are durable. He views the current talent shortage in security as a gap that can only be bridged through aggressive technological advancement.

He prioritizes seeing the present as clearly as possible rather than trying to read tea leaves about the future. He spends his time with practitioners and Chief Information Security Officers (CISOs), the executives responsible for an organization’s data and information security, to understand the ground truth of their daily operations. He believes extraordinary things happen at the convergence of hustle, focus, and humility.

Deep Dive into the Theses

Thesis 1: The Rise of the Technically Sophisticated Buyer

The Opportunity: There is a fundamental shift in who holds the checkbook for security tools. While the buyer may remain the same, the decision maker is moving from traditional IT departments to cybersecurity and engineering teams. These new decision makers are builders rather than more traditional administrators.

The Rationale: As technology becomes more complex, the technical barriers for practitioners have actually risen. For example, spinning up a simple S3 bucket (a public cloud storage service) requires managing 60 or 70 different configurations. To secure this, a lead needs to understand the deep technical implications of every toggle. Security analyst roles, which are often entry-level, are stalling as automation takes over; security engineer roles are the fastest-growing category in the market.

The Conversation: This shift creates a tension between platforms and best-of-breed solutions. While massive players like Palo Alto Networks are acquiring specialized startups like Dig Security to offer a fully-featured platform, Jake observes that sophisticated teams often reject these all-in-one platforms. They want the specific tool that solves a high-stakes problem with the most depth.

However, there is a risk in this trend. Many in venture tend to over-extrapolate how much of the market is actually sprinting toward these advanced tools. In reality, a large majority of the market is still in a crawl phase. Choosing best-of-breed tools requires a level of cybersecurity sophistication that most companies haven’t reached yet; for them, a stable platform is often the more secure, albeit less innovative, choice.

Thesis 2: AI as a Catalyst for Net-New and Reinvention

The Opportunity: AI will have a greater impact on security than the cloud because it demands two things simultaneously: the reinvention of existing tools and the creation of net-new categories.

The Rationale: Cloud computing was essentially an extension of on-premise storage. AI, however, introduces agents that are non-deterministic, meaning they can produce different outputs from the same input. This creates a magnitude of risk that traditional tools cannot handle. For instance, code generation tools can introduce subtle vulnerabilities that a standard code scanner might miss.

The CISO Perspective: This isn’t just an investor theory. When Felicis surveyed CISOs, the consensus was clear: they expect AI to have a far greater impact on security than the cloud ever did. They are currently grappling with how to wrap their heads around these non-deterministic workflows and the brand-new surface area of attack they introduce.

New Attack Vectors: Jake is looking for companies that address new threat vectors like prompt injections, where an attacker manipulates the input of an AI model to bypass security filters, and the evolution of phishing. In the past, phishing was easy to spot due to broken English or grammatical errors. Today, AI allows anyone to generate perfectly written, highly convincing attacks at no cost.

Thesis 3: Shifting Left toward Proactive Defense

The Opportunity: The industry is moving away from the reactive model of debugging toward a proactive model that prevents the bug from existing. This is often referred to as shifting left, or moving code quality protocols earlier into the development lifecycle.

The Rationale: In an era where AI-native attacks move at machine speed, organizations no longer have days or weeks to respond. The standard process currently involves releasing code, scanning it, finding a vulnerability, and creating a Jira ticket for an engineer to fix it later. This relies on too many man-hours and leaves a window of risk open. One hour spent on prevention is worth 15 hours of remediation after a breach.

The “Why Now”: We are roughly three years into the AI reinvention cycle. History shows that the winners of a new era typically emerge a few years after the initial technological shift. Jake points to companies like Corridor as a prime example. Instead of scanning code after it is written, Corridor provides a contextual model of a company’s entire codebase. As an AI agent writes new code, the tool can identify a business logic flaw in real-time before the code is ever deployed; this effectively moves the security barrier left, to the very start of the process.

Conviction and Skepticism

Areas of Skepticism: While some investors are building theses around quantum computing and its ability to break current encryption, Jake remains focused on the immediate compounding risks of today’s AI-driven attacks. He notes that Felicis is focused on “seeing the present as clearly as possible”.

Diminishing Services: Historically, about half of the cybersecurity market has been made up of services; human consultants performing tasks like penetration testing (simulated attacks to find gaps). Jake believes AI is rapidly absorbing this revenue. Companies like Tines (automation) and Terra (AI pen-testing) are replacing billable human hours with high-margin software. This isn’t a total collapse of the services industry; it is a reorganization where software takes over the repetitive, labor-heavy tasks.

Looking Ahead

The Next Frontier

Jake is closely watching the convergence of specialized line items. In the identity sector, the market has evolved from $2 billion companies like Ping Identity to $18 billion giants like Okta. He expects the next generation of identity companies to consolidate even more functionality, creating even larger market opportunities as disparate security needs reorganize into unified, AI-driven solutions.

Final Word

“Cybersecurity is fascinating because it’s a mission-critical function that requires the industry to reinvent itself with every shift. It’s one of the most exciting times ever to be a cyber investor because everything is up for grabs.”

Jon Chu’s entry into the competitive world of funding technological disruption was earned through two decades in the trenches as a hands-on builder, founder, and operator. As a Partner at Khosla Ventures, his deep operational experience gives him a unique vantage point: he invests in the future, not based on market trends, but on a deep, practical understanding of engineering feasibility.

Jon is at the forefront of the firm’s most forward-looking investments on Artificial Intelligence, channeling his technical past to identify the companies that are laying the new foundation for the economy.

The Builder’s Pedigree: From Code Commits to Capital

Jon’s perspective is rooted in a career defined by technical grit and scaling organizations, and his background gives him the ability to connect with founders as a peer, offering an operational empathy few VCs possess.

Scaling Experience: His career began as an early engineer at Palantir, where he learned the rigors of scaling mission-critical systems. He then made the leap to founder, successfully selling KoalityCode to Docker, before running Docker’s enterprise division.

AI in Production: At Opendoor, he held dual roles as Head of Product for Growth and Head of Engineering for Core Machine Learning: Direct, hands-on experience applying ML models to drive a multi-billion dollar business.

Leadership: Most recently, he led engineering organizations at Meta (Facebook) in the cutting-edge fields of Virtual Reality and Machine Learning.

With a B.S. in Computer Science from Carnegie Mellon University, Jon’s path is clear: He is betting on technical excellence because he has lived it.

The Vision: Betting on the AI Infrastructure Beneath the Hype

Jon’s investment thesis moves beyond the surface-level enthusiasm of generative AI to focus on the structural shifts in the enterprise. His influence in the sector is defined by his ability to anticipate not just the power of AI, but the necessary infrastructure of trust and security it demands. He sees the challenges of deepfake technology and fraud not as obstacles, but as significant market opportunities:

“Generative AI enables new deepfake technology that creates new risks and challenges around fraud and trust. Challenges that celebrities, influencers, and brands are not prepared for today. And Loti AI (see below) has world-class technology paired with a category leading product.”

This conviction drives his investment across the AI stack, from core foundation models to the developer tools that empower engineers to build securely and efficiently. He is betting on the companies solving the hard, non-glamorous problems that define true market readiness.

The Portfolio: Foundational Bets Across the AI Stack

Jon’s portfolio reflects this conviction, focusing on the infrastructure: The operating systems, developer tools, and security layers that the entire AI economy will be built upon. Here is a breakdown of his commitment to foundational technologies across the AI value chain:

1. Core Bet: Novel Foundational Models (Sakana.ai)

His investment in Sakana.ai is a prime example of a non-consensus bet on the future of AI.

The Unique Technology: Sakana.ai is a research lab focused on building new foundation models but it is inspired by nature’s principles of collective intelligence and evolution rather than simply scaling up massive, centralized transformers. This approach positions the company to potentially create smaller, more efficient, and specialized models that are better suited for specific enterprise tasks.

Straying from the Crowd: While many VCs flock to fund the latest iteration of large, closed models, Jon is backing a team pursuing a fundamentally different architecture. This reflects Khosla Ventures’ core philosophy of supporting “outlier” technologies with the potential for massive disruption.

2. Developer Infrastructure and Efficiency (Loft / vCluster)

Jon’s backing of Loft (creators of vCluster) showcases his commitment to improving the day-to-day efficiency of developers; a critical bottleneck for scaling machine learning operations or MLOps, the complex process of taking an AI model from a laptop and successfully running it in the real world.

The Foundational Technology: Loft’s vCluster (virtual Kubernetes clusters) allows developers to run virtual, isolated Kubernetes clusters within a single host cluster.

The Strategic Value: This is a pure infrastructure play that directly addresses the pain of resource optimization and dependency management (huge challenges when running large-scale ML training and testing jobs). Jon sees this tooling as essential for any enterprise that needs to run complex AI workloads at scale, making it a critical, non-obvious piece of the AI value chain.

3. The Critical AI Trust Layer (Loti AI)

This investment reflects Jon’s insight that technological disruption always creates a demand for counter-disruption, a theme often missed by application-focused investors.

The Unique Bet: Jon led the Series A funding for Loti AI, a company that develops technology to detect and remove unauthorized likeness content, such as deepfakes and voice simulations.

The Strategic Value: As he stated, the rise of Generative AI creates massive new risks around fraud and trust. Investing in Loti AI is a strategic move to build the necessary regulatory and security scaffolding around the AI future. He sees security and trust protocols as foundational components that must scale alongside the generative models themselves.

The Edge: Why a Builder’s Insight Matters

Jon’s portfolio is characterized by these high-conviction, deep-tech bets that build the underlying capabilities of the AI ecosystem, cementing his position as an investor focused on the engineering rigor required to build a new technical paradigm.

Jon goes beyond writing checks and supports the core components that future developers and enterprises must rely on. His unique background grants him the clarity to evaluate a company’s technical future more deeply than the general market might, actively forging the AI frontier one strategic investment at a time.

More from Uncovered Media

Founders Fund’s Trae Stephens sees a future shaped not just by software, but by industrial-scale, mission-driven manufacturing. In a market where many venture capitalists chase consumer AI and crypto, Trae has become the definitive voice for a philosophy that is deliberately contrarian, philosophically robust, and deeply intertwined with national security: technology that is necessary, even if it is unpopular.

Trae’s career, from the US Intelligence Community to Palantir, and co-founding the defense behemoth Anduril Industries, gives him a unique lens on the nexus of technology, geopolitics, and capital. For him, the biggest opportunity for founders is not to avoid the messiness of hard, physical problems, but to embrace them.

TL;DR

The Factory is the Weapon: In modern geopolitical competition, the ability to rapidly and scalably manufacture software-defined hardware (like autonomous drones and aircraft) is the ultimate form of deterrence. This re-industrialization is a strategic necessity, making large, verticalized production facilities (like Anduril’s Arsenal-1 megafactory) a critical investment for national security.

Reject the Dual-Use Mandate: The requirement that defense technology must have a parallel commercial application hinders innovation. Founders should instead focus laser-like on building the best possible, defense-specific solutions, even if the government is the sole customer. Founders Fund is willing to stomach the single-buyer risk because the payoff—becoming the next major defense prime—is worth it.

The “Feels Bad, Is Good” Quadrant: Trae explains that the most valuable, contrarian ventures often reside in this 2x2 framework (Feels Good/Bad versus Is Good/Bad), embracing missions that may provoke public or ethical discomfort (e.g., defense, mass surveillance) but whose technological success is necessary for a greater, often overlooked, good. It is a call for founders to embrace high-stakes, ethically complex problems.

Founders Fund

Founders Fund is one of Silicon Valley’s most celebrated and controversial venture capital firms. Known for its early, aggressive bets and its philosophical orientation toward making things that matter, the firm’s track record includes initial investments in companies like PayPal, Facebook, SpaceX, Palantir, and OpenAI.

The firm is known for investing in “paradigm-shifting technologies” and has demonstrated a willingness to back companies pursuing massive, seemingly impossible problems. The firm recently raised a $4.6 billion growth fund, signaling its continued interest in deploying concentrated capital into high-conviction opportunities.

Their core philosophy is aligned with the popular ethos of Peter Thiel’s famous quote that “competition is for losers.” The firm is famously anti-consensus, actively avoiding “highly competitive industries” and “consensus categories.” Instead, they prefer to be the first institutional money into a new category, often before the market even recognizes its existence.

This is supported by two main principles:

• Founder-CEO Focus: The firm invests exclusively in founder-led companies, believing the individual founder is the key ingredient in willing a massive company into existence. “Once a company has a professional CEO, we’re out.”

• Technological Risk Tolerance: Founders Fund is willing to write sizable checks into companies that are “a ways out from having a product in market,” demonstrating a deep tolerance for technological risk (a risk that finance-oriented VCs often avoid) in exchange for outsized returns if the technology is realized.

About Trae

Trae’s career path—from the Intelligence Community to the highest echelons of Silicon Valley venture—has uniquely prepared him for his current thesis.

He grew up in Ohio, and the events of September 11, 2001, inspired him to pursue a career in national security; this led him to Georgetown’s School of Foreign Service. After working in government, he joined Palantir Technologies in 2008 as an early employee, where he worked on product and led growth into the defense and intelligence sector. This experience showed him firsthand the disconnect between the speed of Silicon Valley innovation and the antiquated technology used by the government. His path was so focused on this sector that Palantir CEO Alex Karp famously tried to keep Trae away from Peter Thiel for two years to prevent him from being poached.

In 2014, he joined Founders Fund as a Partner, bringing his defense-centric lens to the firm. The pivotal moment came in 2017 when he co-founded Anduril Industries alongside Palmer Luckey, turning his investment thesis into a fully operational company. The goal is to become a “software-defined but hardware-enabled defense prime” and rebuild the US defense industry.

National Security and Moral Conviction

Trae’s conviction is driven by a belief that American technological supremacy is not a given. His early government work, which he describes as inefficient, revealed that the assumption of an impenetrable advantage against adversaries “is just not true.”

This urgency is matched by a strong moral framework. A devout Christian, Trae often discusses his work through the lens of Just War theory, a tradition rooted in the work of St. Augustine. He argues that the goal of incorporating AI and autonomous systems is to achieve greater precision and discrimination on the battlefield; he believes in taking humans out of the “dull, dirty, dangerous jobs” to make defense activities “more just for humanity.”

Investment Philosophy

Trae is a moral contrarian in his investment philosophy. He and Founders Fund seek opportunities that others are structurally or philosophically unable to take.

• Alpha in Non-Consensus: He does not stress about “FOMO” (Fear Of Missing Out). If a category is “hot,” it is likely too late.

• The Origin Story: When evaluating a company, the very first question is about the founder’s origin story. The best companies have incredible, deeply personal reasons for existing, not just a whiteboard-founded idea that is fastest or easiest.

• Concentrated Capital: He advocates for aggressive concentration in winners. For a billion-dollar venture fund, the top five or six companies should represent 40–50% of the deployed capital. If a portfolio company is mediocre, or is getting marked up by a competitor, he is willing to pass on follow-on funding to avoid “stupid single-digit million dollar deals.”

  

  ---

Deep Dive into Investment Theses

Thesis 1: The Factory is the Weapon

Trae contends that the era where the United States was the global manufacturing powerhouse is over, and the consequences for national security are dire. The country’s atrophy in manufacturing—the offshoring of labor and the loss of advanced tooling know-how—has led to a strategic vulnerability.

The central idea is that deterrence is now a function of production speed and scale. In a conflict with a major rival, winning will depend not on the effective but expensive $2.4 million Patriot missile, but on the ability to mass-produce tens of thousands of low-cost, software-defined systems—autonomous aircraft, undersea systems, and munitions—more quickly than any adversary.

The rationale is rooted in the shifting cost calculus of modern conflict:

• End of Expensive Assets: The era of a $15 billion aircraft carrier is over. A single, low-cost hypersonic carrier killer missile (costing only a few million dollars) can threaten that massive, irreplaceable asset.

• A New Industrial Base: The need to rebuild this base is a matter of national prosperity and strategy. Anduril’s construction of the 5 million square foot manufacturing facility Arsenal-1 in Ohio is the tangible proof of this thesis. It is a bet that controlling the supply chain from software design down to solid rocket motor production (verticalization) is the only way to achieve the necessary speed and cost efficiency to create the “magazine inventory advantage” against a major adversary.

This thesis directly challenges the Silicon Valley mindset of “software is king.”

• Capital Intensity and VC: Critics ask how a venture fund with a typical 7–10 year life can reconcile its timeline with the decades of capital commitment required to build a true defense prime. Trae’s reply is through massive follow-on rounds, such as Anduril’s recent $2.5 billion funding, with over $1 billion dedicated to scaling production and building out facilities like shipyards and rocket motor production sites.

• The Speed Muscle: Trae points out that the US has “lost that muscle” for moving quickly. More time has passed between the first flight of the SR-71 Blackbird and today than between the Wright Flyer and the SR-71. Re-industrialization is not just about capacity; it is about regaining that lost speed.

Thesis 2: Rejecting the “Dual-Use” Mandate

This is a direct criticism of the government’s dual-use mandate. The requirement that defense tech must also have a commercial application (e.g., a drone for farmers and the military) is a philosophical and practical hindrance to innovation.

Trae argues that by forcing product designers to compromise on their requirements to serve two masters, the resulting products are made “less effective.” He points out that for specialized systems there “isn’t a dual-use for autonomous weapon stations. You won’t see these systems protecting a college baseball game.”

Founders Fund looks for companies that are willing to forgo the commercial market to deliver a better product, faster for the government. The goal is to encourage founders to pursue hard, defense-specific problems that offer immense, specialized value, rather than chasing diluted, lowest-common-denominator products. The “perfect future state” is one where the Department of Defense (DOD) buys defense-specific items from new players in the same way they buy commercial off-the-shelf products; this bypasses the friction of the dual-use checkbox.

This approach upends standard venture wisdom: avoid single-buyer risk at all costs.

• Vulnerability: By rejecting dual-use, a startup increases its dependency on the government’s highly political and volatile budget cycles. Trae counters that the payoff—becoming a generational defense prime—is worth the risk; the focus should be on getting the government to “pick winners, with meaningful contracts.”

• Commercial Pressure: Critics also argue that the lack of commercial pressure removes the Silicon Valley incentive for cost efficiency and speed. Trae implicitly answers this by building Anduril to be a product company that takes on its own development and production costs upfront; this forces internal efficiency absent in the government’s traditional “cost plus” contracting model.

Thesis 3: The “Feels Bad, Is Good” Quadrant

Trae uses a 2x2 matrix to evaluate the most valuable (and often contrarian) ventures: Feels Good/Bad versus Is Good/Bad. The most compelling opportunities fall into the Feels Bad, Is Good quadrant.

• The Quadrant Defined: This quadrant is for companies like Palantir or Anduril whose missions may provoke public or ethical discomfort, but whose technological success is necessary for a greater good, such as national security, law and order, or saving lives in conflict.

• Embracing Difficulty: The thesis suggests founders should embrace this quadrant and reject the most common trap: ventures that “Feel Good, Are Bad” (e.g., frivolous consumer social apps, or addictive technology designed for hedonism).

This framework provides the moral justification for the firm’s most controversial investments and for working on high-stakes, ethically complex problems in the AI era.

• The Pacifist Paradox: Trae notes, “Pacifism as an idea is only possible inside of a governing system with a monopoly on violence.” Someone must be willing to risk their life to protect the ability of others to be pacifists. This sense of duty and responsibility is what defines the “Is Good” mission.

• The AI Companion Problem: He uses the example of AI companions that affirm the user without disagreement. While this “Feels Good,” he argues it is “Bad” because it avoids the complicated dynamics of real human relationships, ultimately eroding real human connection.

• Talent Allocation: Trae is deeply skeptical of any trend that lures in talented people with low intellectual friction. The ease of building companies that use an LLM API for a highly specific task leads to “whiteboard founding” and a “battle to the death in a really highly consensus category.” Trae says, “If we take all of our level 100 players and we put them on AI slop companies, what does that mean for all of the things that aren’t being done at the same time? I would really like our level 100 wizards to be fabricating semiconductors or, you know, something that’s going to move the needle for humanity.”

  

Looking Ahead

Trae is intensely focused on the final and most difficult hurdle for Anduril: scaling from building thousands of things to tens of thousands of things.

• Production as the Moat: He believes that the next phase of value creation will be in companies that can master design for manufacturing and verticalize the supply chain to cut costs dramatically. This means the capabilities required to produce cars at scale (Tesla) and rockets at scale (SpaceX) are the same capabilities required to become the next major defense prime.

• Valinor Enterprises: His co-founding of Valinor Enterprises points toward a strategy to seed the next generation of focused, product-driven defense players via an incubator launching 10 product companies in one year.

Final Word

While not explicitly stated, his work suggests a core question for the broader tech ecosystem: it’s not just about building better software; it’s about whether we are deploying our ‘level 100 wizards’ on frivolous endeavors or addressing the existential problems that truly move the needle for humanity.

Trae Stephens’ outlook is one of serious, optimistic engagement with the world’s most difficult problems. He offers a framework for founders to choose a “good quest,” defined not by ease or popularity, but by the ultimate long-term necessity of the work:

“I think it’s an opportunity for people to step into a good quest. And so I’m hopeful at least that we’re in a moment of a transition from slop to good questing.”

More from Uncovered Media

Amid the current wave of generative AI investment, the most important question is simple: Where does the value settle? While billions pour into foundation models and chip architecture, Eric Vishria, a general partner at Benchmark, offers a unique perspective.

Vishria frames the core Large Language Models (LLMs) as the fastest depreciating asset in human history. He compares them to the transistor of the 1950s—a crucial, massive enabling technology, but not the ultimate capture point for investors. For Benchmark, the true opportunity lies in the deep, standalone application layer where defensibility is built not on proprietary technology, but on proprietary data loops and deep integration into a user’s mission-critical workflow. This approach shifts the focus from the technology itself to the end-user experience, favoring low barriers to adoption over low barriers to entry. This philosophy, however, comes with a key challenge: a significant portion of early AI revenue, which he calls a “sugar high,” highlights the need for a focus on durable gross margins to build a truly sustainable business.

TL;DR

Models are Depreciating: Foundation models (LLMs) are the fastest depreciating asset in human history. Investment value is shifting to the differentiated products built on top of this commodity.

Value Accrues to “Depth”: The winning AI companies will be “deep applications,” standalone products that solve a highly specific, enduring problem. The moat is the depth of integration into a user’s workflow, which creates high switching costs and makes the product defensible.

Adoption Over Entry: Vishria argues for focusing on low barriers to adoption (Product-Led Growth, or PLG) over low barriers to entry. The new battleground is the customer’s attention and time, so a product wins by allowing users to immediately get value and drive viral adoption.

The Revenue “Sugar High”: A warning that many fast-growing AI apps have revenue that is easy to acquire but is not durable or profitable. This “sugar high” is due to easily replicated features and fragile business models driven by high compute costs, which lead to terrible gross margins.

Benchmark

Benchmark is a venture capital firm known for its disciplined, early-stage approach and a philosophy of active partnership. The firm operates as a flat team of five to seven general partners who share equal economics, a structure designed to reduce politics and encourage collaboration. They make only about eight new investments a year, allowing them to focus on helping their CEOs scale rather than expanding their own organization. Eric Vishria joined the partnership in 2014 as the first addition in six years, and his investments include Amplitude, Confluent (IPO, 2021), Benchling, and Cerebras Systems, as well as Pixie Labs and Bugsnag.

Eric Vishria’s Journey

Vishria’s background provides him with a unique operator-investor perspective, shaped by experiences in investment banking, high-growth startups, and his own company. He graduated from Stanford University at 19. He gained his first entrepreneurial experience as an early employee at Loudcloud and later Opsware, which sold to HP for $1.65 billion in 2007. He then co-founded the social browser RockMelt, which sold to Yahoo in 2013.

This journey gives him a sharp perspective on product-market fit, user adoption, and the eventual commercialization required for his application-layer theses. His experience in the software industry led him to realize that the fundamental wind, the underlying market shift, is more important than the boat (the company). He looks for entrepreneurs who make him “see the world differently” by presenting a unique, plausible, and defensible angle on the market, noting that this “insight” is less a metric and more a unique view of the market. He values founders who are “learning machines,” having the slope, or rate, of learning that will compound past competitors.

Deep Dive into Investment Theses

Thesis 1: As Frontier Models Depreciate Fast, Value Will Accrue to Deep AI Applications

The Opportunity: This thesis is built on the belief that the depreciation of foundational models means value will ultimately be captured by standalone, deep application layer products. Vishria believes that while the initial returns are flowing to the model makers, the enduring venture returns will be made by companies that use this commodity technology to solve highly specific, vertical problems.

The Rationale: The core rationale is that the foundation model layer is the fastest depreciating asset in human history. He compares the LLM to the transistor, calling it a simple “switch” and a massive enabling technology. But just as the value of the transistor has not been captured by the component maker but by the electronics (like cameras and computers) that contain billions of them, the investment value must move up the stack into the differentiated products built on top of the commodity. This forces investors to avoid the platform and focus on defensibility built entirely at the application layer.

The Nuance: This focus on the application layer raises critical questions about what a startup needs to do to establish a moat against horizontal giants like Google or Microsoft:

• The Moat and Deep Integration: The moat for a successful AI application is not the technology itself, but the depth of its integration into a user’s workflow. The successful app must be so integrated that the cost of ripping it out is high, making it defensible even as the underlying models get cheaper.

• The Proprietary Data Loop: For an application to be truly “deep,” it must create a proprietary data loop where usage generates unique, structured data that enhances the product and creates a network effect. Investors must rigorously assess the speed and durability of that loop relative to incumbents that already control the workflow.

• Platform Risk: If a deep vertical application relies on a single foundation model provider (e.g., OpenAI or Anthropic), the startup is at risk of that provider integrating a similar feature at a lower cost. This forces the investment to rely on finding a rare founder archetype that can out-execute and out-learn all others in a rapidly shrinking window.

Thesis 2: Low Barriers to Adoption Over Low Barriers to Entry

The Opportunity: This asserts that the winning products today will prioritize a low barrier to adoption for the end-user, often executed through Product-Led Growth (PLG).

The Rationale: Software is relatively cheap and easy to build, meaning the market is saturated and low barriers to entry no longer guarantee success. Vishria argues that the most important thing that has changed is the lowering of the barriers to adoption. A product wins by having the lowest barriers to adoption, allowing a user to immediately gain value, often for free, without needing a sales team. This viral, bottoms-up adoption is what creates the moat and leads to high-value enterprise sales, as seen in investments like Confluent and Benchling.

The Nuance: The focus on low friction to adoption (through PLG) raises questions about revenue quality and scalability in complex markets:

• The Revenue Paradox: Low adoption barriers often lead to early “free” or “low-price” use. The challenge for investors is identifying the specific metrics and signals in the first 12 to 18 months that ensure this low-friction adoption will successfully convert into sticky, high-value enterprise revenue due to the deep understanding of and integration into client workflows.

• Enterprise Adoption: PLG is great for bottoms-up adoption, but this needs to be reconciled with complex, highly regulated industries (like financial services or health care) where sales will always require a top-down, high-friction security and compliance check.

• Market Saturation: As every new SaaS product adopts PLG, the “low barrier to adoption” itself is becoming commoditized. Furthermore, one of the disadvantages of SaaS is that while it is easy to adopt, it is also easy to adopt a competitor.

Thesis 3: AI Revenue Quality is a ‘Sugar High’

The Opportunity: This thesis is a warning about the quality of revenue many fast-growing AI companies are generating, aiming to distinguish durable growth from temporary excitement.

The Rationale: The “sugar high” refers to revenue that comes easily but is not durable or profitable. Many new AI apps achieve rapid growth by simply wrapping an existing LLM with a nice user interface (UI) for a basic, superficial feature. This growth is fragile because the feature is easily replicated and the companies often suffer from terrible gross margins (e.g., 20% to 30%) because the cost of the underlying model (compute) is high, making the business fundamentally fragile and unprofitable at scale. The goal is to avoid this “sugar high” and find companies with defensible margins and deep utility where the AI is integral, not ornamental.

The Nuance: This thesis forces investors to be extremely disciplined about financial metrics from the very start:

• Pricing Strategy: A fix to the “sugar high” may be pricing. Does the transition from selling a “token” (which leads to a high Cost of Goods Sold and low margin) to selling an “outcome” or a subscription with a fixed price solve the gross margin problem? Charging for business outcomes, such as tickets resolved in customer service, is a potential solution. Additionally, if compute costs continue to drop at a rapid rate, the gross margin problem (the core of the “sugar high” issue) might eventually fix itself, meaning current low margins are tolerable in the short term for the sake of acquiring market share when the costs to those solutions are dramatically smaller in the future.

• Durability: The ultimate resolution to the “sugar high” problem is the durability of the solution, which is only achieved by knowing the clients’ problems intimately enough to solve them. Durable ROI for your client provides the fuel for sustainable revenue.

Looking Ahead

The Next Frontier

The investment in Cerebras highlights a frontier beyond the application layer: specialized, fundamental hardware. Vishria’s early conviction came from realizing that “GPUs actually suck for deep learning training” and that deep learning training is a “communication-bound problem” requiring specialized architecture. This investment was a belief that in an era of massive workloads, the incumbent general-purpose hardware wouldn’t keep up. This shows the next frontier lies in building systems that support 10x bigger models and 100x more data.

Final Word

Eric Vishria brings pragmatism to a market crowded with AI hype. Rather than chasing technology for its own sake, he focuses on the economic model—an operator-first approach that views AI as a powerful but commoditized input, valuable only when it creates durable and profitable businesses.

More from Uncovered Media

Vinod Khosla operates less like a traditional venture capitalist and more like an engineer. As a prominent “techno-optimist,” Vinod refuses to wait for innovation to appear. Instead, he publicly drafts the blueprints for the future and challenges industries to meet his ambitious mandates. His “12 Bold Predictions” exemplify this approach, serving as a roadmap for how technology can drive a more abundant world. Vinod pushes these ideas into the public sphere to instigate and influence the future he envisions.

This philosophy defines Khosla Ventures. The firm operates “beyond capital” by identifying as “instigators.” Their methodology involves pinpointing massive societal problems, such as the “12 major emitter categories” driving climate change, and defining the specific “hardcore technology development” required to solve them. They then actively architect the companies capable of executing the solution.

A central pillar of this vision is Prediction #10: “Fusion and Geothermal Energy Will Replace Fossil Fuels.” While the Earth’s potential heat is limitless, the geothermal industry has historically failed to achieve meaningful scale. That prediction moved closer to reality in October 2025 with a breakthrough by Mazama Energy. The Khosla-backed startup successfully drilled into the hottest rock ever reached by an enhanced geothermal system (EGS); they accessed temperatures of 331°C (629°F) using proprietary techniques.

This event validates a highly specific thesis. Vinod does not merely predict the future; he engineers it. By defining contrarian mandates, such as the necessity of “superhot rock,” he dismisses incrementalism. He deploys a long-term strategy to back entrepreneurs capable of transforming geothermal from “inconsequential” to essential.

The Philosophy of the Instigator

Vinod builds his approach on the conviction stated in Prediction #12: “Entrepreneurs Will Solve Climate Change.” While he acknowledges a role for the public sector, Vinod argues that only radical innovation driven by instigators can deliver the necessary scale. He points to examples like carbon-capturing cement plants, which can double production while cutting emissions, to show how innovation reshapes industries faster than regulation.

This belief system demands a high-risk, “Black Swan” investment model. Khosla Ventures routinely funds projects that conventional investors dismiss as science experiments with extreme risk and long timelines.

Vinod redefines this risk. By framing the climate crisis as an existential threat, he argues that incremental remedies are insufficient. The true risk is not backing radical attempts, but failing to try. As he has stated, “Some things are too important not to try.”

Vinod acts as an architect investor. He identifies technological bottlenecks preventing an industry from scaling and focuses the firm’s resources on solving those exact problems. He provides the public thesis and partners with or incubates the teams required to execute it. This “venture assistance” goes beyond writing a check; it involves shaping the strategy and structure of the solution. He also leverages the power of boldness, arguing that “improbability is what drives progress.” By publicly challenging established norms, he shapes the technological roadmap to attract entrepreneurs willing to challenge the status quo.

The Geothermal Critique: Clearing the Field

Vinod attempts to build the future by dismantling the present. His critique of the existing geothermal industry is deliberate and foundational to his investment thesis. The Mazama Energy announcement opened with a stark dismissal: “Almost all geothermal startups to date have been inconsequential in affecting the energy picture in the US.”

The repeated use of “inconsequential” is a strategic tool. Vinod dismisses conventional geothermal because it is geographically limited, relies on lower temperatures, and cannot scale. In his estimation, the industry is paralyzed by incrementalism; it focuses on optimizing a flawed model rather than removing underlying technological barriers. Vinod ignores niche solutions. His mandate is “impact at tens or hundreds of GW scale.”

This critique serves a strategic purpose. By framing the existing industry as a failure, Vinod rhetorically clears the field. This creates an intellectual vacuum where the central problem is a lack of massive scale. In his view, only his specific approach to hardcore technology development can solve this.

The Mandate: The “Superhot Rock” Thesis

Vinod goes beyond critique to define a non-negotiable set of technological requirements. His thesis focuses not on improving general geothermal energy, but on accessing a new resource: “superhot rock.”

The central requirement is the temperature mandate. Vinod asserts that scalable geothermal requires extreme heat between 400°C and 500°C. The physics of extreme heat changes the economics: at these temperatures, power density increases exponentially. Mazama Energy is designed to “extract up to 10x more power density” than current approaches. This leap transforms geothermal from a geographically constrained resource to a globally available one.

However, this resource must be economically viable. Vinod insists that climate technologies must achieve unsubsidized market competitiveness. They must hit the “Chindia price,” a cost point where developing nations adopt the technology simply because it is the cheapest option. For Mazama, the target is less than 5 cents per kilowatt-hour (kWh) to compete with natural gas.

To execute this thesis, companies must overcome two technological barriers. First is deep drilling. Accessing 400°C+ temperatures requires drilling to depths where conventional technology is too expensive. Second is mastering Enhanced Geothermal Systems (EGS). This technology creates artificial reservoirs where none exist by fracturing hot, dry rock and circulating fluid to extract heat.

The Execution: Mazama Energy and the Integrated Portfolio

The Mazama Energy breakthrough represents the tangible execution of this thesis. At their pilot site in Newberry, Oregon, Mazama created the world’s hottest EGS, reaching 331°C (629°F). This achievement utilized several key innovations, centered on Mazama’s proprietary Thermal Lattice stimulation. This patented process uses advanced hydraulic fracturing techniques to create reservoirs in previously inaccessible rock. Mazama is hitting the targets set by Vinod and explicitly aims for 400°C (750°F) in 2026 to reach the cost competitiveness required for terawatt scale.

Mazama is not an isolated bet. It is the result of a patient strategy by the Khosla Ventures team spanning more than 15 years. The portfolio is not diversified to hedge risk; it is a coordinated, systems-engineered solution designed to solve the specific problems Vinod identified.

• Phase 1 (The Foundation): Beginning around 2008, Khosla Ventures invested in Altarock Energy to prove the science of EGS at the high-heat-flow Newberry site. This was the initial high-risk experiment.

• Phase 2 (The Tool): Recognizing that deep drilling costs were the key barrier to global scale, Khosla helped found and fund Quaise Energy. Quaise is pioneering millimeter-wave drilling technology to access extreme depths economically.

• Phase 3 (The Commercial Vehicle): In 2023, Mazama Energy spun out of Altarock. Incubated by Khosla Ventures, it serves as the vehicle to integrate these advancements and scale the “superhot” EGS model.

This architected approach defines “venture assistance” in action. It solves foundational science, develops necessary tools, and creates a commercial entity to integrate them.

The Catalyst: AI and the Energy Crisis

The timing of the Mazama breakthrough aligns with a new energy crisis. The surge in AI workloads and hyperscale data centers has created an “unprecedented demand for continuous, high-density baseload power.”

Traditional renewables like solar and wind are intermittent. They cannot provide the uninterrupted, 24/7 dispatchable power required by AI. Natural gas and coal are carbon-intensive and geographically limited. This makes the geothermal platform of Mazama an ideal solution: “Always-On, Anywhere, Carbon-Free.” By developing resources at 300°C to 450°C, Mazama can deliver power globally. This allows data centers to operate independently of weather, time of day, or existing grid capacity.

While Vinod remains a proponent of fusion, he notes that “geothermal will happen much faster” as it is closer to commercialization.

Conclusion

The journey from the prediction of Vinod Khosla regarding geothermal energy to the tangible breakthrough at Mazama Energy illustrates the power of his methodology. The impact of Vinod stems from his role as an instigator. He defines the necessary technological breakthrough, dismisses the status quo, and patiently architects a solution through dedicated partners.

The investment of Khosla in superhot geothermal serves as a pillar of his worldview. He is attempting to manufacture an abundant resource from a previously scarce one. By aiming to provide limitless, cheap, dispatchable energy, Vinod works to create a future where the critical inputs for human progress, energy and intelligence, are abundant and nearly free.

More from Uncovered Media

For SV Angel, one of the Bay Area’s most storied firms, success hasn’t come from chasing trends, but from letting the market reveal them.

Instead of trying to predict the future, they listen to it.

Topher Conway, a principal at the firm whose family’s angel investing history dates back to 1993, outlines a philosophy that is less about conviction and more about consolidation. SV Angel’s process is a relentless, high-volume search for pattern recognition, built entirely on a referral system. They do not chase trends; they wait for founders to bring them to their door, signaling a market shift from the ground up.

As Topher notes, their job is not to predict, but to be educated by the entrepreneurs who are spending 110% of their time solving one problem. This bottom-up approach allows them to spot emerging trends, such as the verticalization of AI happening now, not through an internal whitepaper but through the sheer volume of identical, unacquainted ideas flooding their pipeline.

Key Takeaways

Pattern Recognition Over Prediction: SV Angel is not thesis-driven. They identify macro trends by reviewing 40 to 70 referred companies weekly and looking for the same novel idea to appear repeatedly.

The Power of a Small Check: The firm maintains a small Seed fund of $40 million (their growth fund is $330 million) to enable small, checks that encourage collaboration with other VCs.

Intuition is Crucial: The core of their decision-making is a
“gut feeling” or “aha moment” with a founder, a lesson learned early, having made his first investment (Napster) at age 13.

Inflection Point Support: Portfolio management for their large network is scaled by focusing support only on crucial inflection points: early sales/business development, corporate development, and financing.

The Platform: Firm-Level Strategy

The Firm’s Identity

SV Angel’s current fund platform dates to 2009, but its founder-first philosophy has remained consistent for decades. Today the firm typically invests $150,000–$200,000 at the seed stage, backing roughly one new company per week. In recent years, SV Angel has also operated a growth fund to continue supporting select breakout companies beyond the seed stage. All investments are sourced exclusively through referrals from its highly curated 30-year network, reflecting a strategy focused on exceptional founders over specific industries.

SV Angel is focused on the “early stage of company formation” and has participated in funding some of the most iconic technology companies of the past two decades.

The Overarching Philosophy

The firm’s overarching philosophy is defined by alignment and collaboration. By maintaining a small seed fund size (around $40 million), SV Angel can insert small, flexible checks into funding rounds. This approach achieves three strategic goals:

Founder Differentiator: In a market with ballooning check sizes, their smaller, early investment helps them slot in easily.

Investor Collaboration: It keeps the firm non-combative with almost every other VC, fostering strong relationships and increasing their access to high-quality deal flow (referrals).

LP Returns: The belief is that smaller funds tend to generate better returns for Limited Partners.

Alignment

Topher’s focus on high-volume review and deep pattern recognition is a direct continuation and “institutionalization” of the firm’s original angel investing style. His work across both seed and growth funds aligns perfectly with the firm’s strategy of being a flexible, network-driven resource for founders at every stage of their development, from first check through growth financing.

The Navigator: Individual Thinking and Background

The Journey to the Thesis

Topher’s background in venture capital is almost a birthright, with his father starting the family’s investment journey in 1993. As the youngest of three, he was most exposed to the early-stage process. The pivotal moment that shaped his investment lens came at age 13 when his father sought his opinion on an early investment in Napster.

He describes the moment he first used the file-sharing service as an “aha moment.” It was a gut feeling that this was “amazing” and possible on the internet. This formative experience instilled a core belief in trusting one’s own visceral, emotional reaction to a new technology.

The “Why”

This personal experience in the late 90s drives his conviction today: Venture capital is a business of pattern recognition and gut instinct, not just spreadsheets. While the data is richer for growth-stage companies, he maintains that the fundamental question remains the same for both seed and growth investments: “Do I believe in the founders to go and build a massive company?”

Investment Philosophy

Topher’s investment philosophy is centered on being a high-leverage, low-time-commitment resource for founders. The firm is highly concentrated, with approximately 75% of its portfolio in the Bay Area, and 20% in New York, focusing largely on US-based software businesses.

They don’t aim to be in the boardroom or lead a financing round. Instead, they leverage their expanding, decades-old network on behalf of the founder for maximum impact with minimum time commitment.

The Core: Deep Dive into Investment Theses

Thesis 1: The Bottom-Up Trend Detector

The “What” (The Opportunity)

The true opportunity lies not in being the first to announce a thesis, but in having the best signal-to-noise ratio for identifying a trend. SV Angel’s referral system and high review volume (40-70 companies a week) turn the firm itself into a market indicator. They are positioned to see a trend as it is forming across multiple, unacquainted founding teams.

The “Why Now” (The Rationale)

In a market saturated with venture capital, founders are the true innovators. Today, founders are exploring the frontiers of AI technology. The rationale for this approach is that the founders who are fully immersed in solving a specific problem are the best educators on identifying underlying market and technological shifts that make their solution timely.

The “Conversation” (Nuance and Debate)

The internal debate at SV Angel is not whether a trend is happening. Instead, the focus is on how many companies they should index in that space. Topher points to historical instances, such as delivery services (DoorDash, Instacart, etc.), where they purposefully invested in several competing firms to ensure exposure to a significant market shift. The art is determining the size of the overall market and the number of viable competing approaches.

The “Winning Profile” (What They Look For)

They look for founders who can generate that initial “aha moment.” In the current market, this translates to software founders who are already showcasing the verticalization of AI: the new markets and applications that are unlocking as the core technology continues to evolve and develop.

Thesis 2: The Network as Scalable Support

The “What” (The Opportunity)

For a firm with hundreds of companies in its portfolio, the opportunity is to transform that portfolio into a massive, defensible asset (network support). They capitalize on the collective experience of this network to provide high-leverage support for founders.

The “Why Now” (The Rationale)

The growth of the overall venture ecosystem has made fundraising an easier problem to solve, but the challenge of scaling a company remains. The network is most valuable in solving tactical, company-critical problems that a founder can’t solve alone, such as early sales, attracting the best talent, or finding business development opportunities.

The “Conversation” (Nuance and Debate)

The key debate is how to maintain high-quality support without getting dragged into the operational minutiae. The firm manages this by strictly focusing on inflection points: business development (early sales and introductions), corporate development (M&A discussions), and financings.

The “Winning Profile” (What They Look For)

A winning company profile is one that can articulate a need that the SV Angel network can easily solve. For recent investments, this is often the need for early sales, making introductions to older, more established portfolio companies to secure initial crucial customers.

Strategic Focus and Pattern Recognition

Risk and Regulation

The firm’s highly concentrated geographic focus (75% in the Bay Area) and product focus (US-based software businesses) inherently limits exposure to certain international market risks or highly specialized, regulated sectors. The firm’s primary focus is on how they can leverage their network to solve problems that arise, such as connecting founders with immigration counsel for an employee.

Second-Order Effects

Topher notes that their high-volume deal review and pattern-recognition process allows them to see second-order effects in real-time. By observing the immediate verticalization that is happening within AI, they are tracking the consequences of the core technology and where new opportunities are being unlocked.

Looking Ahead

Topher suggests the next frontier is the verticalization of AI and the new markets that are being unlocked as the technology continues to evolve. While they are currently investing in this space, the continuous evolution suggests a rapidly shifting landscape where the new applications are still emerging.

Based on his long-standing view on the importance of the founding team, the provocative question Topher might wish to debate is:

“In a world where technology and capital are increasingly commoditized, how much of a company’s eventual success still comes down to the individual conviction and sheer grit of the founder, and not the underlying market trend?”

More from Uncovered Media

Operational experience serves as the primary lens for identifying opportunity in the current venture landscape. Frédérique Dame, General Partner at GV (Google Ventures), embodies this “operator-investor” model. Dame spent 15 years building consumer and enterprise products, including a pivotal role at Uber. She joined the ride-sharing company as employee 85 and led product and engineering efforts that scaled the organization to over 7,000 people. She helped grow operations from 14 cities to over 400 globally, giving her a distinct understanding of what execution at scale requires.

Dame has invested at GV for over seven years. She applies the disciplined prioritization and infrastructure mindset honed during her operating career to seemingly disparate domains. She focuses on the fragmented landscape of women’s healthcare and the nascent shift toward AI-driven “agentic” commerce. Her core conviction is clear. Systemic inefficiencies, whether in managing menopause or automating social commerce, require infrastructure-level solutions built by founders with deep operational pedigrees. Her strategy focuses on capitalizing on the fragmentation of legacy systems. She moves beyond broad platforms to build specialized, intelligent intermediaries.

The GV Approach

As the venture capital arm of Alphabet Inc. (Google), GV operates as a generalist firm investing in companies from seed to growth stage. The firm organizes its investments into five primary sectors: AI, Consumer, Enterprise, Frontier Tech, and Life Sciences. While GV explores emerging trends, its strategy focuses on backing founders across a wide range of industries rather than targeting specific mature markets.

GV distinguishes itself through its focus on the operator. The firm moves beyond simply funding startups: It backs founders with deep operational pedigrees. Veterans from companies like Stitch Fix, Meta, and Uber possess the tactical experience to navigate complex regulatory and technical landscapes. The scale of capital deployed signals a conviction that the next category-defining companies will build deep, proprietary data moats rather than thin application layers.

Dame specializes in disruptive technologies in the consumer space, healthcare, and emerging AI platforms. She co-leads GV’s Women’s Health team, spearheading investments such as Allara Health, Found, Midi Health, and Oula Health. Her background leading product efforts during periods of significant organizational scaling helps her see what others miss.

From Operator to Investor

Dame forged her investment lens through a career defined by technical expertise and a focus on scalable infrastructure.

Dame grew up in France and moved to the U.S. after earning master’s degrees in telecommunications and spacecraft technology. She wanted to work on products she would actually use, so she switched from enterprise software to consumer roles at Yahoo. While managing social strategies there from 2004 to 2008, she learned how to build at scale. She realized that adding user-generated content, like reviews, meant she also had to build the tools to moderate it.

A pivotal shift occurred in 2012 when she moved to Uber. Although the company already operated in 14 cities, she spent the next four years helping scale operations to 68 countries. She spearheaded strategic programs focused on scaling the global driver workforce and automating internal systems for employee productivity.

This experience taught her the critical lesson of prioritization. When she joined, critical driver information lived in spreadsheets. Her task involved moving operations to automated systems and building the backend infrastructure to automate driver onboarding. When users noted that the early app lacked GPS directions, Dame recognized the strategic necessity of focusing on supply infrastructure before customer-facing features. She noted that adding driving directions helps little if the supply is missing. Without the supply, you have nothing. This infrastructure-first mentality helps define her approach at GV today.

Today, Dame applies that same supply-side logic to a different problem: women’s health. Just as she once identified that an app is useless without drivers, she recognized a massive gap in the medical infrastructure available to women. She observed that many women “suffer in silence” simply because they “are used to how it is.” The data highlighted a clear mismatch between supply and demand. Dame points out that while women live to be 80 on average, they spend “a third of their life” in perimenopause or menopause. Without the right care infrastructure to support women their quality of life is undermined. For Dame, the goal became building the systems necessary so women aren’t just living longer, but “living better longer.”

Thesis 1: Expanding Beyond Fertility

Historically, venture capital investments in women’s health focused mostly on fertility and pregnancy. Frédérique Dame and her team at GV have taken a broader approach, emphasizing that those areas make up only a small part of a woman’s life. Instead, they look for companies that support women through their entire biological lifecycle. This means backing startups that fill specific gaps in care, ranging from chronic conditions to menopause and maternity infrastructure.

This strategy aims to fix the disconnected nature of the current healthcare system. The data shows a clear need for this wider focus. For example, about 57 million women in the U.S. are currently navigating menopause symptoms, yet there is a shortage of doctors trained to treat them. This gap in care affects the economy, resulting in an estimated $1.8 billion in lost wages annually. Additionally, more than one in three women in the U.S. suffer from chronic conditions like PCOS or endometriosis, where care is often scattered and difficult to access.

LinkedIn

GV believes scaling requires integration into the established reimbursement infrastructure. The firm favors specialized virtual care layers over generalist telehealth. They back founders who prioritize insurance integration.

Midi Health: GV led the Series A for Midi, a virtual care clinic for women in midlife. Midi prioritized insurance integration. They positioned themselves as the first menopause-care platform to provide meaningful insurance coverage. This strategy embeds the platform within the healthcare payment stack and turns it into workforce retention technology.

Allara Health: GV led the $10 million Series A for Allara which functions as a virtual “medical home.” It creates a collaborative care platform that pairs patients with doctors and dietitians to manage PCOS and endometriosis. Allara demonstrated rapid traction by going in-network with major insurers in eight states within one year.

Thesis 2: The Agentic Commerce Revolution

Parallel to the shift in healthcare delivery, Dame bets on a fundamental change in commerce architecture. Investments in Nectar Social, Daydream, and Checkmate suggest that the traditional e-commerce stack is becoming obsolete. GV funds “agentic” interfaces and social operating systems that leverage AI to change how consumers find and buy products.

This thesis relies on the behaviors of Gen Z and Gen Alpha consumers. The core insight involves the decline of the traditional marketing funnel. Younger consumers discover products on social media and expect to transact in the moment. Brands often struggle with the volume of interactions. Furthermore, first-party attribution data is critical in a post-cookie advertising world.

Nectar Social: Dame co-led the $10.6 million seed round for Nectar. This platform uses AI agents to automate responses to high-intent interactions. It effectively turns the social media inbox into a sales channel. Nectar solves the attribution problem by connecting specific social interactions to purchase data.

Daydream: GV participated in a $50 million seed round for Daydream. Led by e-commerce veteran Julie Bornstein, Daydream posits that generalist search engines fail for taste-based categories. It leverages Generative AI to understand nuanced, natural language queries. It functions as an intelligent layer that disrupts the standard search paradigm.

Thesis 3: The Infrastructure Imperative

Dame’s operational experience translates into a conviction that foundational infrastructure unlocks scale. This lens extends across her portfolio. She focuses on the invisible infrastructure that enables efficiency and mitigates risk in fragmented markets.

As industries scale, reliance on manual systems becomes a liability. Automation and digitization serve as necessary preconditions for sustainable growth.

TMRW Life Sciences: TMRW provides critical back-end infrastructure for fertility care. As the demand for IVF grows, the manual management of frozen specimens creates risk. TMRW introduces an automated, robotic platform for tracking and storing frozen eggs and embryos. This investment secures the supply chain of biological assets.

Laurel: GV participated in Laurel’s $100 million Series C. Laurel is an “AI time platform” for professional services that automatically captures and categorizes work. Dame notes that Laurel creates the enterprise intelligence layer for knowledge work. By capturing time data, the system makes knowledge work measurable and automatable.

The Power of Vertical Specialization

Dame’s investments remain highly vertical and specialized. Midi solves menopause. Daydream solves fashion search. Laurel solves professional services timekeeping. This implies a skepticism that horizontal, generalist platforms can effectively address complex conditions or nuanced workflows. The conviction is that value lies in deep, specialized integration.

By applying the lessons of rapid scaling to the specialized domains of women’s health and AI-driven commerce, Frédérique Dame helps construct a new operating layer for the economy. In this layer, care is continuous, and commerce is intelligent.

More from Uncovered Media

In an environment saturated with fleeting trends and the hyper-speed narrative of Artificial General Intelligence (AGI), it is easy for founders to mistake momentum for actual progress. This challenge is precisely what Y Combinator (YC) partners Michael Seibel and Dalton Caldwell address in their recent counsel, urging founders to reject the allure of easy wins and the panic-driven pursuit of AGI. Their advice provides a grounding force, arguing that the principles of commitment, patience, and embracing complexity remain the true foundations of building durable, generational companies.

TL;DR

Embrace the Hard Problems: Difficulty is a feature, not a bug, of a valuable solution. If a problem feels too hard, it often means others are avoiding it, creating the potential for a powerful economic moat.

Reject the Guidebook Myth: True startup success is not found by searching for “cheat codes” or a rigid set of best practices (the “MBAification” trap). It is achieved through sustained, non-linear work.

Prioritize Learning: The correct response to the AI era is to strategically and impatiently learn the new tools while maintaining a long-term business focus.

Avoid Short-Term Betting: Guard against the “casino-ification of modern capitalism”: the habit of seeking wealth through quick, short-term bets, by committing to the long-term, hard work necessary to build a durable, value-creating company.

Y Combinator

Founded in 2005 by Paul Graham, Jessica Livingston, and others, Y Combinator is the original and most famous startup accelerator. It is a program that provides early-stage funding, mentorship, and a network to companies in exchange for a small equity stake. YC pioneered the concept of funding startups in batches, applying a mass production technique to early-stage investing that has since been replicated globally. Its mission is to increase the number of successful startups by helping founders master the fundamentals of product development, scaling, and achieving product/market fit. YC’s alumni list includes iconic companies like Airbnb, Dropbox, Stripe, and Coinbase.

The Partners

Both Seibel and Caldwell are quintessential YC partners, having successfully navigated the startup ecosystem as founders themselves before joining the firm’s leadership.

Michael Seibel is a seasoned entrepreneur who co-founded the live-streaming platform Justin.tv, which later pivoted and became Twitch, selling to Amazon for $970 million. Seibel served as the CEO of the YC Startup Accelerator from 2016 to 2024, leading efforts to expand access and promote diversity among founders, notably through programs like Startup School. He currently serves as a Group Partner, continuing to drive YC’s core strategy.

Dalton Caldwell also has significant founder experience, having led the digital music service imeem and co-founded the ad-free social network App.net. As a Partner at YC, Caldwell has advised many of the firm’s most successful companies and is known for his focused, practical advice on navigating product development and avoiding common pitfalls like “tar pit ideas.”

Deep Dive into The Theses

The Anti-Easy Button (Valuable is Hard)

This thesis argues against the cultural tendency among modern founders to treat problem-solving like a simplified test, where maximum difficulty is a signal to pivot rather than double down.

The “What”: The core assertion is that everything that is valuable is hard, but not everything that is hard is valuable. The authors critique founders who, when faced with genuine complexity, abandon the problem entirely, much like a student trained to assume a difficult math problem means they are missing the simple trick. They cite a founder who, upon finding the customer’s biggest challenge, determined it “seemed too hard to solve” and spent months searching for easier, less important problems. True value, they contend, is only secured by grinding through the complexity others avoid.

The “Conversation”: The paradox of startup success is that founders must be both “insanely impatient” to execute rapidly and committed enough to know that “doing great work takes a long time.” The search for an easy answer, they argue, is a trap of “MBAification”—the belief that there is a rigid guidebook or best practice framework that guarantees victory. This mindset fails to recognize that durable advantage, or an economic moat, a sustained competitive barrier, is constructed through the unique, long-term mastery of a difficult problem. The goal is to provide an “Easy Button” for the client, not for the founder to find a shortcut.

The “Why”: When Gusto, formerly ZenPayroll, launched, they didn’t reinvent finance; they simply moved payroll from paper and phone calls to a clean online interface. Users loved it because it allowed them to “forget about it.” Similarly, OpenSea started as a “simple website” that barely worked and relied on third-party wallets. By grinding out a simple use case, they built a massive company.

Historical Context: This thesis is validated by YC’s most successful alumni, particularly those in infrastructure and enterprise. Stripe (YC09), for example, tackled the immensely hard, regulated problem of unifying and simplifying global payment infrastructure. Likewise, early companies that dealt with complex regulated environments, like finance or biotech, or deep technical challenges, like database architecture, faced years of “toil in obscurity.” These founders resisted the temptation to pivot to an easier consumer app, understanding that the difficulty was the very barrier that would prevent competition once their solution finally clicked.

Countering the “AGI or Nothing” Mentality

This thesis is a necessary warning against the emotional and financial overreactions to the perceived imminent arrival of AGI, arguing that market panic can be a greater threat to a startup than the technology itself.

The “What”: The authors caution against the toxic belief that “anything not spent on some sort of AGI pill type life trajectory is a bad call.” Versus waiting for AGI to solve everything for us, it is still very valuable to build companies utilizing current AI capabilities. This is the likeliest path to solving current customer problems. They fear founders are making short-term, panic-driven bets and pivots away from profitable, real-world businesses purely out of a fear of missing out (FOMO) on the perceived AGI wave.

The “Why” This panic is driven by what Caldwell calls the
“casino-ification of modern capitalism.” Young founders have been implicitly trained—via crypto cycles and viral hits—to believe wealth is the result of short-term gambling rather than sustained effort. Seibel challenges founders to list the software companies they respect most. Rarely did any of them “blow up” overnight. Real success, even in the AI era, involves sustained periods of long work.

The “Conversation”: The tension lies in balancing impatience with endurance. Seibel and Caldwell argue that the “AGI era” does not change the fundamental laws of business physics.

Impatience: Founders must be insanely impatient to learn new tools and ship code.

Patience: Founders must accept that building a generational company takes a long time. The correct response to the AI era is not to pivot wildly out of FOMO, but to integrate these new tools into a long-term plan. “Toiling in obscurity,” as Palantir did, is not a sign of failure, it is often a prerequisite for success. Distribution, securing customers, and solving a problem better than anyone else still require long-term focus. The greatest enduring companies of the internet era were built by founders who understood the enduring opportunity, not just the fleeting hype. A founder’s patience and judgment are more valuable than their panic, confirming that even in the AI era, there is still time to do great work.

Why This is Important

The AGI cycle feels distinct from previous shifts, such as the internet or mobile, due to its immediate, dramatic productivity gains and the accompanying existential narrative, which lends an urgency to the call to action. However, the YC partners argue that the fundamental laws of business have not changed. While the tools are new and powerful, the principles of building distribution, securing customers, and solving a problem better than anyone else still require long-term focus. The greatest enduring companies of the internet era were built by founders who understood the enduring opportunity, not just the fleeting hype. A founder’s patience and judgment are more valuable than their panic, confirming that even in the AI era, there is still time to do great work.

The combined theses from Seibel and Caldwell are a defense of foundational principles in a chaotic market. For founders, they represent a required course correction against the impulse to chase speed and simplicity. The advice provides clarity in an age of confusion: don’t look for the guidebook or cheat code. Generational companies are not built by those seeking to minimize effort, but by those who find a profoundly difficult problem and commit to solving it with sustained, focused effort. By rejecting the easy path and the panic button, founders can build a valuable company that will survive the market’s cycles and capitalize on the long-term, structural opportunities presented by technological shifts like AI.

More Uncovered Media

In the landscape of startup acceleration, Y Combinator (YC) has long been a dominant force in the “launch fast and iterate” model. But in a market flooded with B2B SaaS, one of YC’s recent Request for Startups (RFS) signals a major, deliberate shift.

Diana Hu at YC is particularly interested in hard tech. She is known for her conviction that the real opportunity is in using AI to solve fundamental problems in the physical world. She isn’t just predicting improvement; she’s identifying a new area for high-risk, massive-reward investing.

TL;DR

From SaaS to Science: YC’s RFS, championed by Hu, signals a major appetite shift toward hard tech. They are specifically looking for LLMs to be applied to hard sciences, such as discovering drugs or new materials, not just writing emails.

The Infrastructure Bottleneck: A second thesis, co-authored with Dalton Caldwell, targets the “single biggest bottleneck in the current AI boom: physical infrastructure.”

The “Lights Out” Datacenter: The bet is on non-software innovation. The ask is for “’lights out’ autonomous datacenters managed by robots and software”, a bet on “concrete and electricity, not just code.”

Challenging the YC Model: These theses represent a move toward “high-risk, massive-reward investing” with “very long feedback loops” (e.g., FDA approval), which directly challenges the traditional YC model of “launch fast and iterate.”

The Capital Hurdle: This new focus raises questions about capital, asking if a startup can realistically compete with hyperscalers if traditional early-stage raise amounts do not significantly change.

The Platform: Y Combinator

The Firm’s Identity: Y Combinator (YC) is a highly successful startup accelerator, built on a philosophy of rapid iteration and scaling. Its traditional model, “launch fast and iterate”, has defined a generation of software-first companies.

The Overarching Philosophy: YC’s core philosophy is expressed through its intensive batch program and its widely read Request for Startups (RFS). This RFS acts as a powerful signal to founders, directing entrepreneurial energy toward problems YC believes are most critical. While historically focused on software, the firm’s philosophy is evolving to support new categories of innovation.

Alignment: Diana Hu’s work aligns directly with YC’s new strategic mandate. As a champion of the “AI for Scientific Advancement” RFS, her focus is not a side project; it is a direct execution of the firm’s strategic shift. This new focus on hard tech and deep domain expertise aligns YC’s platform with a new class of high-risk, high-reward founders.

The Navigator: Diana Hu

The Journey to the Thesis: Diana Hu’s perspective is not that of a traditional investor; it’s that of a deep tech operator and founder. Her journey was forged by a “geeky learning obsession”, leading her to an MS in Electrical and Computer Engineering from Carnegie Mellon with a focus on computer vision and machine learning.

Her path wasn’t a straight line to VC. She was an operator at Intel and OnCue TV (sold to Verizon). The pull to build led her to co-found and serve as CTO of Escher Reality (YC S17), an AR backend company. This YC founder experience gave her deep operator empathy.

After Niantic acquired Escher Reality, she led their AR platform as Director of Engineering. Her return to YC as a Group Partner completes the cycle, as she is now a technical founder guiding the next generation of hard tech.

The “Why”: Hu’s drive is her constant curiosity to learn” She sees her work not just as investing, but as an “outlet of my exploration and curiosity”. She views technologies like AR, and by extension the hard tech in her RFS, as the “next technology evolution” after desktop and mobile. As an immigrant founder herself, her “why” is deeply connected to helping other founders navigate the high-stakes, technically complex journey she has already completed.

Investment Philosophy: Hu’s philosophy is founder-focused and expertise-driven. Having been a technical CTO and YC founder, she invests “in places where I feel I can truly contribute and be helpful to founders.”

She is not a generalist. She focuses on spaces where she has deep expertise, such as “computer vision, machine learning or VR,” which perfectly mirrors her “AI for Science” and “Autonomous Datacenter” theses. She views investing as a long-term journey and looks for founders with a growth-oriented mindset whom she will be “happy to continuously support.”

The Core: Deep Dive into Investment Theses

This is where we explore the high-conviction ideas shaping YC’s updated worldview. Two key theses emerge: the reorganization of scientific discovery by AI and the robotic future of the datacenters that power it.

Thesis 1: AI for Scientific Advancement

The “What” (The Opportunity): This thesis moves beyond generic AI hype to target hard sciences. Hu is “specifically looking for LLMs applied to discover drugs or new materials”. This is a deliberate “move toward high-risk, massive-reward investing.”

The “Why Now” (The Rationale): The thesis taps into the current excitement around AlphaFold and similar breakthroughs, suggesting that AI is finally mature enough to tackle fundamental biology and physics problems.

The “Winning Profile” (What They Look For): This thesis requires founders with deep domain expertise. A prime example from the current YC batch is Zeon Systems, a company that lets you describe an experiment in plain text which its robot arms then execute. The bet is that this approach could “drop [the price of scientific research] 90%.”

The “Conversation” (Nuance and Debate): This bold thesis brings up critical challenge questions:

• The YC Model Challenge: “Scientific advancement historically has very long feedback loops (e.g., FDA approval)”. How does this mesh with the traditional YC model of “launch fast and iterate?”

• The Tool Mismatch Challenge: Are LLMs actually the right tool for all scientific advancement, or is this “forcing a language paradigm onto hard physics/bio problems that need different architectures?”

Thesis 2: Rethinking Datacenters with Software and Robotics

The “What” (The Opportunity): This thesis, co-authored with Dalton Caldwell, “addresses the single biggest bottleneck in the current AI boom: physical infrastructure.” It is a “very ‘non-software’ software thesis.” YC is not just asking for better management tools; they are “asking for ‘lights out’ autonomous datacenters managed by robots and software.”

The “Why Now” (The Rationale): The thesis is anchored by the fact that “energy and space constraints begin to hit major AI players.” It’s a “bet that the sheer demand for compute will force innovation in concrete and electricity, not just code.”

The “Conversation” (Nuance and Debate): This strategy surfaces two critical VC questions:

• The Hyperscaler Challenge: Can a startup realistically compete in datacenter innovation when giants like Google, Amazon, and Microsoft are spending tens of billions annually on this exact problem?

• The Capital Challenge: Is this a capital-intensive play that requires far more than the traditional early-stage checks to get off the ground?

The Bottom Line

Diana Hu’s investment theses provide insight into YC’s evolution. While much of the market remains focused on iterating SaaS, her RFS focus remains clearly on deep domain expertise and hard tech.

Her bets on AI for Science and autonomous datacenters demonstrate a conviction that the next wave of value won’t come from novelty, but from the hard work of integrating AI into real-world, physical workflows. It’s a pragmatic, accelerator-first perspective that sees AI not just as a tool for writing emails, but as a tool of discovery.

More from Uncovered Media

In the landscape of AI investment, few voices are as specific and testable as Cathy Gao, a Partner at Sapphire Ventures. In a market flooded by generic ‘AI in healthcare’ talk and fears that ‘Microsoft will just add that feature,’ Cathy’s perspective on Sapphire’s investment strategy cuts through the noise. Helping to lead the firm’s approach in these verticals, she emphasizes that the real opportunity is not in competing with horizontal giants but in going ‘insanely deep’ into messy, specific industry workflows.

TL;DR

Vertical AI is the Moat: Startups will not win by competing with Microsoft’s horizontal copilots. They’ll win by going “insanely deep” into complex verticals (like tax law) that big tech cannot or will not touch. Depth is the new defensibility.

The ‘10x Provider’ is Coming: The healthcare stack is being rebuilt around three new AI layers: data activation, administrative automation, and, ultimately, tools to make a physician 10x more effective.

Observations on Voice AI: Reflecting on market observations, Cathy noted the potential for Voice AI to become the default patient engagement channel within the next 12-18 months. This indicates a possible acceleration of healthcare’s typical tech adoption cycle.

The 2025 Test: ROI is the Hurdle: Looking ahead, Cathy observed that while AI budgets appeared poised to expand, the new challenge would be “turning hype into actual proof”. She noted potential “last mile problems” like governance and privacy that could hinder true autonomous agent adoption.

Sapphire Ventures

The Firm’s Identity: Sapphire Ventures is a global B2B software venture capital firm managing over $11 billion in assets. The firm’s primary investment team, where Cathy is a Partner, is known for its expansion-stage focus, typically targeting Series B through IPO companies that have already achieved scale (e.g., $10M+ in Annual Recurring Revenue). Their track record includes over 80 exits including 30 IPOs, with historical wins like Square (Block), LinkedIn, DocuSign, Looker, and MuleSoft.

The Overarching Philosophy: Sapphire’s core philosophy is its “Portfolio Growth” platform. This is not just a value-add; it is a dedicated, in-house team of operators whose job is to help companies scale. In practice, this means providing portfolio companies with strategic customer introductions to Global 2000 companies, access to a network of over 7,000 executives for key hires, and advisory on international expansion. This hands-on, operational support is a key part of their model and has earned the firm an 82 CEO Net Promoter Score (NPS) from its portfolio companies.

Alignment: Cathy’s work is a direct execution of the firm’s central strategy. Sapphire Ventures is aiming to invest over $1B in enterprise AI, with a deep focus on ‘vertical SaaS’ and ‘health tech.’ Cathy is helping to lead Sapphire’s strategy across these verticals. Consequently, the firm’s investments in Blue J (Vertical AI for tax) and EliseAI (Vertical AI for housing and healthcare) are not side projects; they are the playbook.

The Founder’s Question: “Am I too early for Sapphire?” While Sapphire’s official target is expansion-stage ($10M+ ARR), the firm has made exceptions for an exceptional thesis fit. The primary focus is on growth, typically leading B2B rounds in the $25M to $60M range and advising on $30M+ Series C rounds. However, Sapphire may lead an earlier round if the conviction is high, as demonstrated when the firm led the Series A for Jeeva.ai.

Cathy Gao’s Journey

The Journey: Cathy’s perspective is uniquely shaped by a career spent on all sides of the table. As the daughter of a software engineer and a small business owner, she was “immersed in technology and entrepreneurship from a very early age”.

Her path was not a straight line to VC. She started in finance, building fundamental models as an analyst in Blackstone’s M&A group and later in growth equity at TA Associates. But she felt the pull to build, moving to San Francisco to become an operator at the fast-growing Series B startup Gusto, where she worked across finance, strategy, and product. This experience gave her the “operator” empathy that now defines her. She returned to investing, first at AXA Venture Partners and then joining Sapphire in 2019.

The “Why”: Her dual background in finance and operations gives her a passion for “deeply understanding business models” while also “working hand-in-hand with entrepreneurs to help scale companies”. She describes investing as the “best job in the world” because it pays you “to learn something new every day”. This drive is built on her stated core values: conviction, partnership, and passion.

Investment Philosophy: Cathy is known for being “incredibly founder-focused,” acting as a “thought partner and a true champion through the ups and downs”. When evaluating opportunities, she values an investor’s ability to be “creative and proactive in gaining even a slight edge to build conviction,” whether through a unique network or by “connecting seemingly unrelated dots”.

Deep Dive into The Theses

This is where we explore the high-conviction ideas shaping Cathy’s worldview. Two strong theses emerge: the reorganization of healthcare by AI and the defensible moat of deep, vertical-specific AI.

Thesis 1: The “10x Provider” & The AI Healthcare Stack

The “What” (The Opportunity): This thesis moves beyond generic hype to map a specific, three-layer stack that will redefine care delivery. Cathy’s model includes:

• Data Activation: Unlocking the vast, siloed data within health systems.

• Admin Automation: Deploying AI to manage the heavy administrative burden.

• The “10x Provider”: Creating AI-powered tools that augment a physician’s capabilities, aiming to make them 10x more effective and efficient.

The “Why Now” (The Rationale): The thesis is anchored by a strong, time-bound prediction. Cathy predicts that Voice AI will become the default patient engagement channel within the next 12-18 months. This ambitious timeline suggests a belief that the technology is finally mature enough to break healthcare’s notorious adoption bottlenecks.

The Conversation (Nuance and Debate): This bold thesis brings up critical challenge questions:

• The Adoption Cycle Challenge: Healthcare is famously slow to adopt new “default” channels, citing complex regulations like HIPAA and older patient demographics. What makes Voice AI so compelling that it can overcome this inertia in just 12-18 months?

• The Burnout Challenge: The “10x provider” concept is powerful, but what if the tools fail? There’s a significant risk that new AI tools could increase alert fatigue (the overwhelming number of digital notifications) worsening physician burnout rather than reducing it.

Thesis 2: Vertical AI Beats Horizontal Incumbents

The “What” (The Opportunity): This thesis directly counters the prevailing market fear that “Microsoft will just add that feature”. It concedes that large, horizontal players like Microsoft and Google will win the general-purpose AI game (e.g., basic copilots). The real venture-scale opportunity is for startups that go “insanely deep” into specific verticals.

The “Winning Profile” (What They Look For): These startups win by integrating into the messy, high-stakes, and specific workflows—like tax law or property management—that horizontal players simply won’t or can’t touch. This is a high-conviction bet that depth is a sufficient moat against big tech. As horizontal AI becomes a commodity, the value shifts to applications with deep domain expertise. Sapphire is backing this thesis with significant capital, as seen in their investments in Blue J (Tax) and EliseAI (Housing).

The Conversation (Nuance and Debate): This strategy surfaces two classic, hard-nosed VC questions:

The TAM Question: Does extreme verticalization limit the Total Addressable Market (TAM) too much? If the AI is wildly successful at automating away the very complexity that justifies its high price, does the market opportunity shrink, making venture-scale returns impossible?

The “Services” Trap: Are these vertical AI companies just “services businesses” disguised as high-margin software? High-stakes fields like tax law often require a heavy human-in-the-loop (HITL) to manage exceptions and provide assurance, which can destroy software margins and scalability.

Revisiting Cathy Gao’s 2025 Predictions

Its not often that we get to see a VC’s time-bound predictions play out in real-time. But thanks to an interview Cathy Gao gave on Bloomberg Technology in early January 2025, titled “The Evolution of AI Agents”, we can do just that.

As of November 2025, we analyzed three primary discussion points from that interview. It is important to note that statements made regarding predictions in January 2025 are a reflection of Cathy’s observations at that time only, and that Sapphire Ventures has no influence on those statements / viewpoints. The status and analysis reflect the observations of Uncovered Media.

Source:The GenAI Divide STATE OF AI IN BUSINESS 2025 (MIT)

Why This is Important

Cathy Gao’s investment theses provide a clear playbook for navigating the hype. While much of the market chases horizontal platforms or general-purpose AI, her focus remains firmly on vertical specialization and proof. The firm’s investments in vertical-specific AI and the measurable ‘10x provider’ demonstrate a conviction that the next wave of value won’t come from novelty, but from the hard work of integrating into messy, real-world workflows. It is a pragmatic, operator-first perspective that sees AI not as a universal fix, but as an effective tool for solving specific, high-stakes business problems.

More from Uncovered Media

John Locke

The Uncovered Crew is diving into the thesis of John Locke, a partner at Accel. In a venture landscape flooded with AI-first wrappers and horizontal apps, Locke is making massive, high-conviction bets in a completely different direction.

He’s backing companies that are building AI-native systems of record for complex, less exciting enterprise verticals. Locke’s thesis is that the real, defensible value isn’t in the AI model itself, but in using AI to rebuild the core infrastructure and workflow of entire industries—like legal and finance—from the ground up.

And when he finds a company that proves this thesis, he’s willing to ignore traditional fundraising timelines and invest with speed and conviction.

TL;DR: Theses and Key Quotes

Bet on Plumbing, Not Paint
“The real opportunity is not in thin AI features but in building new, AI-native systems of record (like an ERP) that replace legacy giants.”

Vertical Wins
“Target industries with high complexity, massive spend, and ancient incumbent software, like legal (Filevine) and corporate finance (Campfire).”

Conviction > Convention
When market demand is overwhelming and a thesis is being proven, act with speed. Locke led a Series A and co-led a Series B for Campfire just 12 weeks apart.

On the AI System of Record
“Filevine is tackling a problem that others are missing by developing a system of record where AI is seamlessly woven into every function...”

On Overwhelming Market Signal
“That’s always a great proxy for us... The demand is quite a bit higher than what we even anticipated.”

The Platform: Accel

To understand Locke’s thesis, you have to understand Accel’s platform. The firm has a long history of backing founders with “grit and scrappiness” who have proven their model, often by bootstrapping to significant scale before taking on major capital.

This “bootstrapped lineage” includes giants like Atlassian and Qualtrics. This philosophy shapes a worldview that values substance over hype and looks for a strong, proven foundation for long-term success. Venture capital is seen as an accelerant for a proven engine, not fuel to find a match. Locke’s recent investments are a modern-day extension of this philosophy: find a company solving a real, expensive problem and then provide overwhelming firepower to win the market.

The Navigator: John Locke’s Philosophy

John Locke’s investment philosophy is defined by a search for structural disruption. Based on his public writing and recent investments, his focus is not on what’s merely popular but on what is foundational.

His thesis isn’t about “AI” in isolation. He always pairs it with structural terms: “system of record” (for Filevine) and “ERP for the future” (for Campfire). His investments are not bets on a specific AI model; they are bets on full-stack business-model disruption that is enabled by AI.

He looks for signals that cut through the noise. While many VCs are chasing horizontal AI applications, Locke’s trigger is seeing intense, real-world market pull from sophisticated buyers. When other Accel portfolio companies are clamoring to use a product, that’s the signal.

Thesis 1: The “AI-Native” System of Record

The core of Locke’s thesis is that standalone AI tools are inefficient. He argues that “fragmented tools create silos” and that the true winner in any enterprise vertical will be the core operating system where all the work already happens.

The opportunity isn’t to sell an AI feature to a law firm; it’s to replace the law firm’s entire legacy operating system with one built from the ground up with AI at its core. He targets industries like legal and finance because they have high complexity, massive spend, and are trapped on ancient incumbent software (think NetSuite or SAP).

This is a bet on the “plumbing” over the “paint.”

Supporting Investments:

Why This Is Important

This thesis provides a clear map for navigating the AI hype cycle. It suggests that the long-term, defensible venture outcomes will not be “wrappers” but full-stack, vertical-specific platforms. By focusing on the system of record, Locke is betting on companies that can achieve true platform lock-in, embedding themselves into the mandatory, core workflows of an entire industry.

Enjoying Thesis Uncovered? Subscribe for the latest content.

Subscribe now

Thesis 2: Aggressive, Rapid Follow-On (The Conviction Thesis)

When a thesis is being proven correct faster than anticipated, Locke’s actions show he is willing to throw out traditional fundraising timelines and invest with extreme aggression.

The signal is overwhelming market demand. His thesis is that in this scenario, the primary risk is not capital or valuation; the primary risk is speed.

The primary evidence is his investment in Campfire.

• June 2025: Accel leads the Series A.

• October 2025: Accel co-leads the Series B just 12 weeks later.

This is a massive, public signal of conviction. Locke’s own rationale provides the “why”: the demand from sophisticated buyers (including other Accel portfolio companies) was “immense” and “quite a bit higher” than even they had anticipated.

He didn’t wait the typical 18-24 months. He moved to capitalize the company to meet the opportunity now.

Why This Is Important

This thesis is a high-conviction approach to venture. It demonstrates a willingness to double down—aggressively and publicly—when a company’s real-world traction outpaces the original model. It suggests that when a company has clearly hit a nerve and is solving a deep, expensive problem, the right move is to end the race for capital early and focus all energy on execution and market capture.

More from Uncovered Media

Vinod Khosla is not a recent convert to the transformative potential of Artificial Intelligence. As a co-founder of Sun Microsystems during the microprocessor boom and an early investor in the foundational technologies of the internet (such as Netscape and Juniper Networks), Khosla has spent four decades identifying and capitalizing on era-defining technological shifts. His perspective on AI is therefore not merely theoretical but grounded in a career spent at the inflection points of the digital age.

This analysis tracks the evolution of Khosla’s thesis on AI over a pivotal decade, from 2014 to 2024. It is a journey that moves from early economic warnings to urgent geopolitical priorities, culminating in a grand vision of a “post-scarcity” future. A post-scarcity world is one where technology makes essentials like energy, food, and expertise so cheap and abundant they are practically free. Khosla’s thinking has not been static; it has adapted significantly to accelerating technological capabilities and shifting global dynamics.

In 2014, he issued a warning shot, arguing that AI would drive unprecedented abundance alongside a massive gap between the rich and the poor. This challenged the widely accepted economic theories that technology always creates more jobs than it destroys. By 2017, his focus sharpened on an immediate threat: the weaponization of AI by adversarial nation-states (specifically China and Russia). He argued that the greatest danger was not sentient machines (conscious AI that can feel or perceive things), but losing the AI race.

His current thesis, articulated in 2024, synthesizes these concerns into a comprehensive worldview. Khosla argues that AI represents a “difference in kind” from previous technologies—multiplying the human brain much as engines multiplied muscle power. This potential for near-infinite brainpower necessitates an aggressive acceleration of AI development to secure a utopian future and ensure the prevalence of democratic values, while simultaneously demanding a radical rethinking of capitalism and social policy to manage the inevitable disruption.

The Khosla Doctrine on AI: Acceleration, Abundance, and Adaptation

The Utopian Goal: Achieving Post-Scarcity AI promises unparalleled abundance and a post-scarcity economy. However, achieving this utopia requires implementing radical policy interventions, such as Universal Basic Income (UBI) (an unconditional sum of money provided by the government to all citizens) and “empathetic capitalism” (Khosla’s term for a system focused on reducing disparity, not just profit), to manage massive job displacement.

The Evolution: From Disparity Concerns to Existential Risks Khosla’s thesis evolved from recognizing that AI fundamentally replaces human judgment (2014), to prioritizing the immediate threat of AI-driven cyberwarfare and geopolitical imbalance (2017), to advocating for aggressive acceleration as an existential necessity (2024).

The Geopolitical Necessity: Winning the AI Race While initially focused on general economic disparity, Khosla increasingly prioritized the existential threat of adversarial nations dominating AI. He argues that “slowing down the development of AI could be disastrous for democracies and the greatest risk we could possibly take.”

Defining the Shift: Why AI is Fundamentally Different

“Difference in Kind”: AI is not just another tool; it is the intellectual parallel to the steam engine, multiplying brainpower rather than just augmenting human labor.

“This Time Is Different”: The historical precedent that technology revolutions create new jobs may not hold true when technology surpasses human intelligence and judgment.

Deflationary Growth: A specific economic concept where prices fall (deflation) while well-being and consumption increase (growth). AI may drive down costs so significantly that traditional GDP (Gross Domestic Product) metrics may become distorted or even decline.

The 80/80 Prediction: Automating Expertise and Ending “Survival Servitude” A future where Khosla predicts “80% of 80% of all jobs” will be automated, expertise (doctors, tutors, engineers) is near-free globally, and humanity is liberated from “survival servitude” (being “enslaved” to the need to work just to survive) to redefine its purpose.

Note: Vinod uses the “80% of 80%” phrasing to make his prediction specific: he believes AI will impact the breadth of 80% of job categories, and will automate the depth of 80% of the work within those jobs, for a total of 64% of all work tasks.

Khosla Ventures: The Engine Room for “Reinventing Societal Infrastructure”

source: slides for AI: Dystopia or Utopia? (May 2025)

Khosla Ventures (KV), the venture capital firm founded by Vinod Khosla, is the platform from which these theses emerge and the vehicle through which they are actioned. The firm’s identity is rooted in backing disruptive innovations—technologies that completely change or overthrow existing industries—that promise significant societal impact. This spans information technology, sustainable energy, healthcare, and advanced materials. The firm is structured to take long-term, often high-risk bets on technologies capable of “reinventing societal infrastructure.”

Khosla’s background at Sun Microsystems and his history of investing in foundational technology platforms inform KV’s core philosophy: that profound technological shifts, while disruptive, are the primary drivers of human progress. The firm operates with a belief that the “possible” must be actively brought into existence. Khosla’s evolving AI thesis aligns directly with this mandate, viewing AI not just as a sector for investment, but as the defining technology of the next several decades, capable of reshaping every vertical the firm touches.

The “Technology Possibilist”: Khosla’s Philosophy and Motivations

Vinod Khosla operates as a strategist identifying macro-level transformations. His perspective is deeply informed by his engineering background and his experience navigating previous technological revolutions. He identifies as an “unapologetic capitalist” and a “technology possibilist”—someone animated by the “boundless potential of what could be.”

Khosla’s motivations appear twofold. First, a profound belief in technology’s potential to solve very difficult or (nearly) impossible human problems, such as providing free healthcare and education globally. He frequently cites the “bottom half of the planet”—the 4 billion people who stand to gain the most from AI-driven abundance, contrasting this potential with the “ivory tower” concerns of Western academics. Second, a pragmatic concern for the preservation of Western democratic values. He views the AI race with China as a battle for the future of global influence, arguing that the nation leading in AI will “anoint the world’s political system.”

His analytical approach is characterized by a willingness to challenge economic orthodoxy and anticipate sudden breaks from previous trends rather than predict the future by assuming existing trends will continue the past. He favors causality over correlation, frequently dismissing economists who rely on historical trends. As he quoted Karl Marx in 2014, “when the train of history hits a curve, the intellectuals fall off.”

Mapping the Journey: A Three-Stage Evolution

Vinod’s analysis of AI has evolved significantly over the past decade, adapting to the accelerating pace of technological development and the changing geopolitical landscape.

2014: The Economic Warning—“This Time Is Different”

Thesis: “The Next Technology Revolution Will Drive Abundance And Income Disparity”
Link to Source

The Argument: Devaluing Human Judgment Khosla argued that machine learning was rapidly progressing toward surpassing human judgment, decision-making capability, and potentially even creativity. This would dramatically increase productivity and economic abundance.

The Catalyst: Big Data and Complex Decisions The emergence of big data and progress in complex tasks previously thought impossible for computers (e.g., driving a car). Khosla noted examples in his portfolio aiming to replace functions ranging from farm workers to cardiologists and legal researchers.

The Significance: Challenging the Luddite Fallacy This essay was significant for its stark warning: “This time could be different.” Khosla challenged the “Luddite fallacy.” The Luddites were 19th-century workers who destroyed machinery, fearing it would take their jobs. The “fallacy,” according to mainstream economics, is the belief that technology destroys jobs; the orthodoxy holds that it always creates new ones. Khosla argued that historically, technology augmented human capability. But if machines become superior in intelligence and judgment, human labor will be fundamentally devalued. Education and retraining, he contended, would not be sufficient solutions for the majority (the bottom 80%) of the workforce.

2017: The Geopolitical Pivot—“Scary for the Right Reasons”

Thesis: “AI: Scary for the Right Reasons”
Link to Source

The Argument: The Immediate Threat of Weaponized AI Khosla shifted the focus from long-term economic disruption—which he estimated was still a decade or more away—to immediate national security threats. He argued that the public discourse (led by figures like Elon Musk and Stephen Hawking) was wrongly focused on dystopian “overlord AIs,” while ignoring the immediate danger: “AI in the hands of governments and/or bad actors used to push self-interested agendas.”

The Catalyst: Cyberwarfare and Authoritarian Investment The context included visible interference in political systems (Russian interference in the US election), large-scale cyber hacks (Equifax), and the aggressive, focused investment in AI by China and Russia for nationalistic advantage.

The Significance: Reframing the AI Arms Race Khosla highlighted the vulnerability of the West. He noted that the open publication policies of Western companies (Google, Facebook) and universities created a “one-way flow of technology.” Furthermore, he argued that authoritarian regimes held an advantage due to their disregard for data privacy. The AI race was reframed as an “economic war,” potentially worse than the nuclear arms race, necessitating immediate countermeasures.

2024: The Utopian Imperative—Acceleration as Survival

Thesis: “AI: Dystopia or Utopia?”
Link to Source, annual meeting talk, slides

The Argument: AI as Infinite Brainpower A grand synthesis arguing that AI is a “difference in kind,” multiplying the human brain itself. This promises a “post-scarcity economy” or “consumer utopia” where material limitations are eliminated and expertise is near-free.

The Catalyst: Advanced Capabilities and the “Doomer” Narrative The rapid advancements in AI capabilities (LLMs, robotics) and the rise of the “Doomer” narrative (the story that AI will lead to catastrophe or the end of humanity), which Khosla aggressively seeks to dispel and reframe.

The Significance: The Forced Choice and Empathetic Capitalism This stage integrates the economic warnings of 2014 and the geopolitical urgency of 2017. The core argument is that the risk of falling behind China far outweighs the risks of AI development (sentient or otherwise). Therefore, regulation that slows progress is the greatest danger. The solution to the 2014 disparity problem is now clearly defined: “empathetic capitalism,” UBI, and leveraging the massive abundance AI will create.

The 2024 Synthesis: A Blueprint for the Post-Scarcity World

Khosla’s 2024 thesis is a passionate argument for accelerating AI development as a moral and existential imperative, combining a utopian vision with pragmatic policy recommendations to navigate the transition.

The Intellectual Engine: Multiplying the Human Brain

The core concept is that AI is the intellectual parallel to the steam engine. Just as engines allowed humans to use external energy (coal, oil) to augment physical output, AI allows for a “near-infinite expansion of brain power.” This transcends human capacity, leading to a future where the vast majority of jobs can be done by AI, often better and more consistently. The ultimate goal is a post-scarcity society where work is a choice, not a necessity, potentially ushering in a 3-day workweek. This democratization of expertise promises near-free AI tutors and physicians globally.

The Forced Choice: The West vs. Authoritarian AI

Khosla directly addresses the dystopian views of “pessimists and doomers.” He argues their focus on the risks of sentient AI—while real—is misplaced, “myopic, alarmist, and actually harmful” because it ignores the far greater risk: losing the AI race to nefarious nation-states. He frames the situation as a forced choice: “Are you ready to trust Xi and his Putin-like appendages for the equitable distribution of one of the world’s most powerful technologies? That would be dystopian.” Slowing down, he argues, is not an option; the “cat is already out of the bag,” and technological leadership is an “existential priority worthy of wartime mobilization.”

Beyond GDP: The New Economics of Abundance

The nuance in Khosla’s argument lies in his vision for the new economy and the redefinition of human purpose. He anticipates “AI-led deflationary growth.” As the cost of labor and expertise approaches zero, the prices of goods and services will plummet. This could lead to a paradox where nominal GDP (GDP measured in current prices, without adjusting for inflation or deflation) decreases, but consumption and overall well-being increase dramatically. Traditional GDP measures, Khosla argues, will become a “distortion of prosperity.”

Furthermore, AI offers the chance to liberate humans from “survival servitude.” He argues that the dystopian narrative is “cognitively lazy” and often comes from those insulated from existential pressures. By eliminating the need to work for survival, AI could increase our “humanness,” allowing people to pursue passions, creativity, and exploration.

The Dual Strategy: Acceleration Meets Adaptation

Khosla’s thesis translates into a dual strategy:

1. Technological Acceleration: Aggressive investment in AI development across all sectors (healthcare, education, resource discovery, bipedal robotics). This includes investment in AI safety research, but vehemently opposes premature regulation that would slow progress relative to China.

2. Socio-Economic Adaptation: Acknowledging that the transition over the next 10-25 years will be “very messy” due to the speed of change (much faster than the agrarian revolution), Khosla advocates for structural changes to capitalism. Recognizing that “Western capitalism is by permission of democracy,” he calls for prioritizing income disparity reduction alongside efficiency (”empathetic capitalism”) and implementing UBI funded by AI-driven abundance.

The Strategic Playbook: Navigating the Transition

source: slides for AI: Dystopia or Utopia? (May 2025)

Vinod Khosla’s decade-long analysis of AI forms a coherent strategic playbook that has evolved from cautionary observation to urgent advocacy. The arc of his thinking reflects a consistent belief that AI is a fundamentally different technology that will disrupt the foundations of labor and economics.

What has changed is the prioritization of threats and the clarity of the solutions. In 2014, the threat was abstract economic disparity and the solution was a tentative call for structural change. By 2017, the threat was concrete—geopolitical imbalance and cyberwarfare. In 2024, the threat is the potential loss of a utopian future and the failure of democratic values, and the solution is an aggressive combination of technological acceleration and radical policy adaptation.

The overall trajectory is a movement from what the economic impact will be (disparity), to who the adversaries are (China/Russia), to how we must manage the transition to achieve the utopian outcome. Khosla’s evolution demonstrates that navigating the AI revolution requires not only aggressive technological advancement but an equally aggressive willingness to reshape the socio-economic foundations of society, ensuring that the transition to a post-scarcity world is guided by deliberate, democratic choices rather than fear or inaction.

More from Uncovered Media

We recently dug into the thinking of Pat Grady and Sonya Huang, key partners at Sequoia Capital and leaders in the firm’s AI practice. In the world of venture, Sequoia has been a central character in the AI story, most notably as an early, pivotal backer of OpenAI.

But what’s compelling isn’t a single thesis. It’s the evolution of one. Starting in 2022, Pat, Sonya, and the Sequoia team have publicly mapped their thinking on Generative AI, almost like an annual report on the state of the revolution.

Their journey tracks the market’s own whiplash: from the initial “better, faster, cheaper” hype of 2022, to a sober “prove your value” correction in 2023, to the rise of “agentic reasoning” in 2024.

Now, they are articulating the next great shift, moving beyond simple text or images. They call it “AI Synesthesia”—a new era where models are natively multimodal, able to connect different types of data (text, images, sound) as if they were all the same format. This isn’t just a technical upgrade; they argue it’s a fundamental “shift in the cognitive interface” between humans and machines, one that finally unlocks intelligence trapped in the wrong modality.

TL;DR: Theses and Key Quotes

• The “AI Synesthesia” Era is Here: The next frontier is not just AI that handles text, image, code, and audio, but AI that thinks in a unified way across all of them. It’s about “unlocking intelligence that’s been latent, unexpressed, or trapped in the wrong modality”.

• From Hype to Utility: The AI market has rapidly matured. “Act 1” was the initial hype. “Act 2” was the correction, focusing on solving “human problems end-to-end” and proving value.

• The Rise of “Thinking Slow”: The latest shift is to an “Agentic Reasoning Era”. This moves AI from “thinking fast” (instant pre-trained responses) to “thinking slow” (not just matching patterns, like a simple chatbot, but is actively “thinking”) to solve complex, real-world problems.

• Moats are Workflows, Not Just Data: Sequoia’s initial 2022 thesis on “data flywheels” was corrected by 2023. They observed that the most durable moats are not in the data itself, but in “customers” and embedding AI into “workflows and user networks”.

• Raising the Floor and the Ceiling: The ultimate goal of AI Synesthesia is to “raise both the floor and ceiling of human capability”. The floor rises as specialized skills (like coding) become accessible to all; the ceiling rises as experts gain new ways to express complex ideas.

The Platform: Sequoia Capital

To understand this thesis, you have to understand the platform it’s invested from. Founded in 1972 by Don Valentine, Sequoia Capital is arguably the defining venture firm of the modern technology era. Its portfolio reads like a history of tech itself, from Apple and Atari to Google, Cisco, PayPal, Nvidia, Airbnb, Stripe, and YouTube.

The firm’s philosophy is famously captured in its tagline: “We help the daring build legendary companies, from idea to IPO and beyond”. This isn’t just marketing copy; it defines their strategy.

• Founder-Obsessed: Sequoia invests in people first, focusing on the founder’s vision and culture.

• Long-Term Partners: The firm sees itself as a “partner for the long term,” famously forbidding the use of terms like “deal” or “exit” internally.

• Team Sport: Grady and other partners emphasize that VC at Sequoia is a “team sport,” not a game of individual showmanship.

• Mission-Driven LPs: A core, motivating part of Sequoia’s identity is that it invests primarily on behalf of nonprofits, universities, and foundations.

As key partners in the firm’s growth-stage and AI investments (co-leading the OpenAI investment), Grady and Huang are perfectly aligned with this mandate: find the next “daring” founder who will build a “legendary” company to define the next 50 years.

The Navigators: How Grady and Huang Think

As the co-authors of Sequoia’s key AI theses and co-hosts of the “Training Data” podcast, Pat Grady and Sonya Huang have become public-facing navigators for the firm’s AI strategy.

As a self-described “Silicon Valley kid at heart” and the daughter of immigrant engineers, Sonya Huang’s perspective was shaped by growing up in the “gravity well of innovation” of the 1990s. Her technical background, which includes an AB in Economics with minors in Computer Science and Statistics/Machine Learning from Princeton, gives her a deeply technical lens, stating, “I’m much more comfortable reading a technical paper or talking to a researcher than going to an industry conference”.

Grady’s background, for instance, is not typical Silicon Valley. He moved from construction jobs in high school to a high-volume inside sales role—making “50 dials a day”—before joining Sequoia in 2007. This experience informs a deep, personal connection to Sequoia’s mission. He notes that his own scholarship to Boston College was funded, in part, by proceeds from the very endowments Sequoia now invests for.

“If we screw up, people lose out on scholarships and cancer research,” Grady has said. “Now that’s motivation”.

This drives the team’s investment philosophy, which believes “culture is everything”. They view founders as the “heroes” of the startup journey and Sequoia’s role as that of a long-term, supportive partner on that “epic journey”. This view is critical to understanding why their theses aren’t just market-timing, but attempts to map long-term, tectonic shifts.

The Core: The Evolution of an AI Thesis

Grady, Huang, and the Sequoia team didn’t just publish one thesis; they’ve been publishing an evolving roadmap. This timeline shows how their thinking has tracked the industry’s own maturation.

2022 Thesis: “Generative AI: A Creative New World”

• The “What” (The Opportunity): This was the foundational “hyperscale” thesis. It declared GenAI a profound platform shift that would empower “billions” of knowledge and creative workers. The focus was on making original work “better, faster, and cheaper”.

• The “Why Now”: A combination of cheaper compute and new techniques (like diffusion models) was unlocking a “creative new world” across text, code, images, and soon video and 3D.

• The Water Cooler: This was the shot heard ‘round the Valley—the moment a top-tier firm declared GenAI was not a toy, but the next major wave of computing.

2023 Thesis: “Generative AI’s Act Two”

• The “What” (The Opportunity): After the “Cambrian explosion” of “Act 1,” the market needed to mature. “Act 2,” they argued, would be defined “from the customer-back”. The focus shifted from models as the product to models as a piece of a full-stack solution that solves “human problems end-to-end”.

• The “Why Now”: The initial hype led to poor retention. GenAI’s biggest problem was no longer discovery; it was “proving value”.

• The Water Cooler (The Correction): This was a critical update. The team explicitly revisited their 2022 thesis and stated that the real, durable moats were not just “data flywheels.” Instead, moats were being built in “customers” and, more specifically, in “workflows and user networks”. This is where companies like Harvey (which Grady led the investment in) thrive: by deeply embedding AI into the specific workflow of a specific industry.

2024 Thesis: “Generative AI’s Act o1”

• The “What” (The Opportunity): This thesis, titled “Act o1”), declared the “Agentic Reasoning Era”. As the foundational layer consolidated around a few scaled players (OpenAI, Anthropic, Google, etc.), the new opportunity is in reasoning.

• The “Why Now”: The shift is from AI “thinking fast” (pre-trained, instant pattern-matching) to “thinking slow” (using compute at inference time, not just delivering a pre-cooked answer learned during training, for deliberate reasoning and problem-solving).

• The Water Cooler: General-purpose models are not enough. The “messy real world” requires customly designed, cognitive architectures and domain-specific reasoning. This is the layer where startups can build massive, defensible businesses on top of the big platforms.

2025 Thesis: “On AI Synesthesia”

This brings us to today. “On AI Synesthesia” is the culmination of the previous three theses. It takes the consolidated models (Act o1) and the need for end-to-end solutions (Act Two) and defines the new human-computer interface that will power them.

• The “What” (The Opportunity): The thesis identifies the shift to natively multimodal models. This is not about “bolting on” an image generator to a text-based LLM. It’s about models that are built from the ground up to understand text, image, code, video, and audio in a “unified latent space”. This unified approach makes the creative process “more controllable” and allows generation to happen “all together” rather than in separate, loosely-coupled systems. The team argues this is a profound “shift in the cognitive interface” between people and computers.

• The “Why Now” (The Rationale): For most of computing history, intelligence flowed through a “narrow channel: text in, text out”. This created a power dynamic: if you were a good writer or coder, you could command the machine. But, as the article notes, “intelligence isn’t limited to words. Some think visually, rhythmically or spatially”. This new class of multimodal AI finally “bridges these gaps” and “unlocks intelligence that’s been latent, unexpressed, or trapped in the wrong modality”.

• The “Water Cooler” (Nuance and Debate): This is AI’s “synesthesia moment,” named for the neurological trait where one sensory experience (like hearing a note) triggers another (like seeing a color). The team points to composer Alexander Scriabin, who “experienced music as color” and could “fluently transition across sensory modalities”. “What was once rare neurological wiring,” the thesis states, “is now becoming our shared digital capability”. In this new era, “creativity becomes translation”. An idea is no longer trapped in the mind of a person who can’t code, draw, or write; it can be “fluidly transferable”.

• The Winning Profile (What They Look For): The team is looking for companies that build for this new “mental operating system”. These winners won’t just be tools; they will be partners that “raise both the floor and ceiling of human capability”.

• Raising the Floor: This means specialized skills become accessible to everyone. “Visual thinkers can speak in paragraphs. Coders can sketch in pixels. Writers can prototype products”.

• Raising the Ceiling: This means experts are given new superpowers. “The barriers between disciplines, roles and modes of thought begin to dissolve”. The winning companies will be those that understand this fundamental shift: “Your strengths are no longer limited to the format you were trained in. Your ideas are no longer trapped in the mode you happen to be best at”.

From Hype to Integration: A Playbook for a High-Speed Market

The nearly four-year journey mapped out by Grady, Huang, and the Sequoia team is a playbook in itself. It demonstrates a thesis that is not static, but a living, breathing framework designed to adapt at the same speed as the technology it tracks.

In 2022, the call was about potential—a “creative new world” of “better, faster, cheaper”. But by 2023, the market delivered a harsh lesson in retention, proving that hype doesn’t pay the bills. Sequoia adapted its thesis publicly, shifting the focus from novelty to value. The moat, they corrected, wasn’t the model; it was the “workflow” and solving “human problems end-to-end”.

This pragmatism paved the way for the next layers of opportunity. The 2024 “Agentic Reasoning” thesis identified the “thinking slow” layer not as a consolidated commodity, but as the critical engine for real-world problem-solving. And 2025’s “Synesthesia” is the human-centric payoff: a new, fluid interface that unlocks all forms of intelligence, not just the verbal ones.

What this evolution shows is a move from what (the generative model) to how (the reasoning agent) to who (the human user, in all their multimodal intelligence). In an industry where change is measured in months, not years, the Sequoia team’s public roadmap suggests the most durable investment strategy is one of relentless adaptation, a willingness to correct course, and a constant search for the human value inside the machine.

More from Uncovered Media

Shreyas Garg (IVP)

Find him on LinkedIn

The Uncovered Crew recently sat down with Shreyas Garg, an investor at the tenured venture firm IVP, to discuss the rapidly evolving landscapes of consumer AI and the enduring utility of specific crypto applications.

In the landscape of venture capital, few firms possess the historical perspective of IVP. Founded in 1980, IVP has been doing venture investing since the Atari and Walkman were the best tech you could buy. This enduring presence gives the firm a unique perspective, having guided market leaders through countless cycles and storms.

IVP maintains a high bar for its next set of crypto investments. Each year, the firm invests in just a dozen breakout founders ready to scale from millions to hundreds of millions in revenue and expand from one market to many. With a remarkable track record—130+ IPOs out of 400 investments, including giants like Netflix, Twitter, Snap, Discord, and Coinbase—IVP helps ambitious founders defy limits, command industries and cement their place at the top.

This commitment continues with the launch of their 18th fund in 2024 (yes, you read that correctly). “We’re launching IVP’s 18th fund at an intense— and promising — time for the industry,” the firm noted regarding the $1.6B raise.

“We are experiencing a sweeping technological shift with AI at a time when capital markets are resetting. From our experience, these periods create massive opportunities for companies with product-market fit, organic momentum, and the right team.” [IVP]

Our goal with Thesis Uncovered is not to capture every possible area our guests would invest in. Instead, our goal is to capture in-the-moment thinking and conversation, like what you would hear around the water cooler or at the hipster coffee shop down the street. Shreyas offered deep insights into two rapidly evolving sectors, emphasizing the need for genuine utility and sustainable business models in both arenas.

TL;DR: Theses and Key Quotes

Utility Over Speculation in Crypto

IVP maintains a high bar for crypto investments, focusing strictly on utility rather than speculation, with stablecoins emerging as the most compelling use case for store of value and movement of money.

“The bar we wanted to hold ourselves to was to make an investment that was tied to a utility related to crypto. We thought stablecoins were one of the most interesting ways people were using crypto, which we thought was a very clear utility.”

The Imperative of Invisible Tech

For mass adoption, the underlying crypto infrastructure must be obfuscated. The value proposition should focus on the benefit (cheaper, faster money movement), not the technology.

“I think the key to getting adoption is honestly just making this less and less obvious that it’s crypto-related. I as the end user shouldn’t really need to understand and think about crypto to get utility out of them.”

The AI Monetization Shift

AI applications, with costs closely tied to LLM usage, are shifting monetization models from flat subscriptions to usage-based pricing, better aligning cost with value.

“In a world of AI where your costs are more correlated to the value the user gets. These apps are now charging more on a usage basis, and therefore the value someone gets is [...] more proportional to value (they) pay.”

Retention is King in Consumer AI

Despite the top-line traction many AI companies are seeing, long-term success hinges on demonstrable, retentive consumer behavior.

“...looking for retentive businesses, I think, matters even more, uh, nowadays, just given that I think a lot of great companies are getting a lot of like top line traction...”

The Proactive Paradigm Shift

The next major UX shift in AI will be moving from reactive models (asking ChatGPT a question) to proactive agents that anticipate needs and automate workflows autonomously.

“I think right now, AI is very reactive versus proactive. And I’m very excited for proactive AI.”

“It’s not like me asking for something or knowing I want something. It’s like, Hey, like, and I can predict what I want and what I need and just deliver it to me.”

How Shreyas Thinks: The Pragmatic Search for Utility

Shreyas approaches emerging technologies with a blend of excitement and grounded pragmatism. His thinking is characterized by a relentless focus on utility. While many investors get caught up in the hype or the technical novelty of a sector—be it Web3 or generative AI—Shreyas consistently drills down to the core question: what real-world problem does this solve, and is the solution demonstrably better than the status quo?

This utility-first approach is evident in his view on crypto. Since IVP had a lot of success with Coinbase, Shreyas is less excited about additional crypto investments related purely to speculation on token prices. He emphasizes discipline and focus, candidly stating, “We have not found a thesis in Web3 outside of stablecoins that we’re very, very excited about.”

Furthermore, Shreyas values what he calls “magical consumer experiences”—that feeling when you use a product and realize, “this is special.” It’s a qualitative metric he describes as “you know it when you see it.”

But he quickly pivots from magic to the underlying business fundamentals necessary to sustain it. In the AI space, he is wary of rapid growth fueled by novelty rather than genuine retention. He recognizes that the benchmarks for success in venture are extremely high; the initial traction is not the finish line.

“A great venture outcome in consumer is probably close to a billion of run rate revenue,” he noted. “So even if you’re like, get really fast, a hundred million of ARR... you still gotta go from a hundred to a billion.” This long-term view forces a rigorous examination of retention dynamics and monetization models early on.

Guest Background and Notable Investments

Born in India and raised in Northern Virginia, Shreyas was exposed to technology early, starting to code at 14. His interest always spanned both technology and business, leading him to study computer science at Princeton. A chance opportunity to shadow a VC during his freshman year solidified his career path. “I just kind of fell in love with what he did for a living,” Shreyas recalled.

He took a strategic route into venture, starting in investment banking at Morgan Stanley before joining IVP about four years ago. After two years in the San Francisco office, he helped launch IVP’s London office, spending 18 months building it out before relocating back to San Francisco.

His notable work includes investments in companies like Traceable AI, an API security firm. He also highlighted Cradle, a company that “builds a protein model for pharma companies that better optimize drug development,” using AI to iterate through protein sequences and speed up the drug development process—a concrete example of applying AI to a core, high-value workflow in an established industry.

Thesis 1: The Utility-Driven Case for Crypto (Stablecoins)

Despite the volatility and speculative nature that has characterized much of the crypto market, IVP remains interested in the sector, provided investments meet a high bar for utility. For Shreyas, the most compelling area that clears this bar today is stablecoins.

“We thought stablecoins was one of the most interesting ways people were using crypto that we thought was very clear utility,” Shreyas emphasized. He identifies two primary applications where stablecoins offer significant advantages over traditional financial systems.

1. Store of Value in Volatile Economies

The first major use case is providing a stable store of value, typically pegged to the US dollar, for individuals in emerging markets with highly inflationary or volatile local currencies.

“One is just like store of value,” Shreyas explained. “You can kind of see the benefit for people, especially in emerging countries, who have more volatile currencies.” When local currency value changes day to day, “your purchasing power fluctuates a lot. And that can be quite rough.” By allowing people to save in a currency less prone to devaluation, “stablecoins can provide you a way to save money in a more stable fashion. Therefore, your purchasing power remains intact.”

2. Seamless Movement of Money

The second application is disrupting the slow, expensive rails of traditional fiat currency movement.

“Two is just like movement of money,” he said. “Obviously like moving like fiat currency has, you have to go to the banks, it takes days. It’s like not the cleanest process in the world.”

Shreyas shared a personal anecdote from his time in London. “That’s actually how I paid my rent in London while I was living there, because I got paid in US dollars, had to spend in pounds,” he said, utilizing apps to move money seamlessly. “Just moving money around more cheaply and faster will lead to, I think, an interesting company being built on top of the rails.”

The Path to Adoption: Invisible Infrastructure

While the utility is clear, Shreyas believes mass adoption hinges on making the underlying crypto infrastructure invisible to the end user. The technology must be obfuscated by a clear, compelling value proposition.

“I think the key to getting adoption is honestly just making this less and less obvious that it’s crypto-related,” he argued. “I as the end user shouldn’t really need to understand and think about crypto to get utility out of them. I should just be able to say, this is the value prop and the actual rails can be somewhat obfuscated.”

The marketing should not focus on the technology; it should focus on the outcome: “I don’t think [the marketing] needs to be about stablecoins, [it] should be move money cheaper, move money faster, save money in like US dollars and not like your local currency.”

Why This is Important

The focus on stablecoins represents a maturation of the crypto industry, moving beyond speculation toward solving fundamental financial problems. It addresses the very real pain points of currency devaluation and cross-border transaction costs. If executed correctly—with an emphasis on user experience and regulatory compliance (Shreyas noted IVP’s preference for working with regulators)—this thesis points toward massive opportunities for financial inclusion globally, potentially disrupting legacy financial institutions by providing a faster, cheaper, and more accessible alternative.

Thesis 2: The Search for “Magical” Consumer AI

The explosion of generative AI has led to a proliferation of new applications, and Shreyas is highly optimistic about the potential. “I think AI will change a lot of how we just interact with the world as consumers,” he said, looking for “AI native” versions of existing categories and entirely new applications.

He is actively searching for those “magical consumer experiences.” However, magic alone is insufficient. Shreyas applies a rigorous lens to consumer AI, focusing on defensibility, retention, and evolving monetization models.

The Retention Imperative and Monetization Shift

A key challenge in the current AI landscape is distinguishing between genuine product-market fit and novelty-driven growth. “Looking for like retentive businesses, I think matters even more, uh, nowadays,” Shreyas noted, observing that many fast-growing companies may not have the “best like core retention dynamics.”

Furthermore, AI introduces a new cost dynamic. Unlike traditional SaaS where marginal costs are low, AI applications have significant compute costs (LLM charges) directly correlated with usage. This is driving a shift away from flat subscription models.

Shreyas used Spotify as a counterexample: a heavy user and a light user pay the same price despite deriving vastly different value. AI changes this equation.

“In a world of AI where your costs are more correlated to the value the user gets, just like you have to pay the LLM charge. These apps are now charging more on a usage-based model, and therefore the value someone gets is closer to the value [they] pay.”

He is looking for evidence that users are willing to bear these costs, particularly the “whale cost” of heavy usage. “If I get a ton of value out of it, like, am I paying more and more? And like, are those users also sticking around?”

The Next Frontier: Proactive AI

Perhaps one of the most exciting shifts Shreyas anticipates is the move from reactive to proactive AI. Today’s tools require user initiation—you have to go to ChatGPT and ask a question. The future lies in AI that anticipates needs and acts autonomously.

“I think right now, AI is very reactive versus proactive. And I’m very excited for proactive AI,” he said.

He illustrated this with his own workflow at IVP. Currently, he might manually request a research report before a meeting. The proactive version would operate without prompting.

“But there’s no reason why an AI shouldn’t just proactively look through my calendar, go through all my first meetings, and just generate a research report for me, send it to my email a day before every meeting.”

This is the essence of the shift: “It’s not me asking for something or knowing I want something. It can predict what I want and what I need and deliver it to me.” This concept extends to optimizing workflows automatically: “Literally just having an AI watch my workflow data and be like, hey, can automate this, can automate that, and just does it for you. That flavor of company could be quite exciting.”

Why This is Important

The shift toward proactive AI represents a fundamental change in human-computer interaction, moving beyond tools to true agents that reduce cognitive load and automate complex tasks. Furthermore, Shreyas’s emphasis on retention and usage-based monetization addresses the critical need for sustainable business models in the AI era. As the initial hype subsides, the companies that succeed will be those that deliver continuous, measurable value directly aligned with what users are willing to pay.

More from Uncovered Media