Key Moments
Bill Maris: How Google Could Crush AI Competitors, Why Small Funds Win, and AI's Atari Stage
All-In PodcastWant to know something specific about what's covered?
We've already dissected every moment. Ask and we will deliver (with timestamps).
Key Moments
Small VC funds targeting higher returns are outperforming large funds, but founders taking large checks at inflated valuations risk misaligned incentives.
Key Insights
Small VC funds (under $750 million) returned an average of 4.76x, while funds larger than $1 billion averaged 2.42x, with 95% of top-decile performers being below the $750 million mark.
Google Ventures, under Bill Maris's leadership, aimed for top-quartile returns and achieved an estimated 4.1x return when compared to publicly available data for top-quartile VC returns.
Founders accepting large checks from giant VC funds at significantly inflated valuations (e.g., $250 million for 1-5% of a company) may face misaligned incentives, even if it seems advantageous initially.
Google's potential strategy of drastically cutting AI token costs (e.g., by 80%) could create immense pressure on competitors like OpenAI and Anthropic, potentially making their business models unsustainable.
AI is in its 'Atari Stage,' analogous to early video games, and is expected to rapidly evolve into a mature, photorealistic, and ambient computing experience within the next five years, driven by underlying platform technologies.
The US is experiencing a 'brain drain' in scientific talent due to an 'anti-science vibe,' gutting of research institutions like NIH and CDC, and restrictive immigration policies, pushing researchers to countries like China.
The entrepreneur's 'insane' glimpse of the future
Bill Maris's entrepreneurial journey began with a profound moment of inspiration in 1997, while working on Wall Street with a neuroscience degree. Discovering a server in his office closet, he realized the potential for a business centered around web hosting and data centers. This led him to quit his job and start a company from his apartment in Vermont, living in one room while his servers occupied another. The conditions were so rudimentary that water would freeze on his desk in winter, and he once had to tar his own roof during a thunderstorm to protect the equipment. This experience, which others likely viewed as 'insane,' taught him that seeing the future often requires a degree of unconventional thinking and willingness to take significant personal risks. He illustrates this with the example of someone live-streaming an inauguration on a laptop in 2009, a seemingly odd behavior at the time that foreshadowed the ubiquity of personal device content creation.
Leveraging data science for Google Ventures
In 2007, Maris was tasked with building Google Ventures (GV). His approach was deeply rooted in data. He collected vast amounts of venture data and, despite initial resistance from Google to use the term 'AI' (opting for 'machine learning'), applied it to design an ideal portfolio construction through millions of simulations and back-testing. The goal was to determine optimal fund size and investment strategies. This data-driven methodology, though initially perceived as 'crazy,' proved successful. During the period of 2009-2018, GV's estimated returns were around 4.1x, significantly outperforming standard top-quartile VC returns of that era. His core lesson drawn from this is a strong conviction: 'Don't bet against computer science.' He asserts that applying the right computer science principles at the right time to the right problem invariably leads to correct outcomes, even in the face of initial skepticism.
The mathematical advantage of smaller venture funds
Maris's fourth lesson challenges conventional VC wisdom: smaller funds outperform larger ones. He states this is not an opinion but a mathematical reality, backed by data showing funds under $750 million averaging 4.76x returns, compared to 2.42x for funds over $1 billion. Funds below $750 million accounted for 95% of top-decile performers, with significant return compression occurring above $750 million. The reasoning is straightforward: a $500 million fund needing to return $1.5 billion (3x) requires $5 billion in exits. In contrast, a $7 billion fund needs to return $21 billion, which can exceed total annual venture exit value. This mathematical constraint forces larger funds to seek much larger, rarer exits, increasing risk. Furthermore, smaller funds allow for greater focus and more dedicated attention to founders, which Maris believes is crucial for successful venture investing. He personally runs Section 32 with six funds averaging $400 million, all performing in the top decile.
AI's disruptive potential and the 'Atari Stage'
Maris likens the current state of AI to the 'Atari Stage' of the gaming industry. Early games like Zork were brittle and text-based, a far cry from modern photorealistic, immersive experiences. He predicts AI will undergo a similar, accelerated evolution in the next five years, moving from command-line interactions to ambient computing. The key drivers won't just be larger models but the underlying platform technologies – controllers, physics engines, and GPUs – that enable richer interactions. While many debate the doomsday or utopian scenarios for AI, Maris suggests the reality will likely be less extreme. His investment thesis focuses on these foundational platform components that will power the next generation of AI applications, rather than solely on the largest foundational models themselves.
Google's strategic pricing power against AI competitors
A significant concern for AI competitors like OpenAI and Anthropic is Google's market position and pricing strategy. Maris outlines a scenario where Google could drastically undercut competitors by reducing AI token costs by, for example, 80%. If a business can access a 'basically identical product' from Google for 80% less, they have a strong incentive to switch. This creates immense pressure on companies that are already burning significant investor cash and have relatively low revenues (e.g., a $1 trillion spend commitment on $60 billion in revenue). Such a move by Google would be a strategic weapon, aimed at capturing market share and building an install base, potentially at the expense of competitors' viability. This intense competition, coupled with companies staying private longer, raises questions about who ultimately benefits.
The disconnect between public benefit claims and investor returns
Maris criticizes companies that use 'public benefit' language while concentrating wealth creation within a select group of investors. He argues that those who genuinely aim to benefit humanity should go public sooner, allowing broader participation. The current trend of companies staying private for extended periods, then potentially making massive IPOs, means that ordinary investors via 401ks and ETFs might become the 'bagholders' of overpriced assets. They are forced to buy into companies at peak valuations without having had the opportunity to participate in the early, high-growth stages. This practice, he contends, leads to wealth creation for the already wealthy and leaves average retirement accounts exposed to significant risk.
Challenges in life sciences and the flight of scientific talent
While Maris has a long-standing interest in life sciences, noting his involvement with Calico and investments in Flatiron and Verily, he acknowledges the specialized nature of therapeutic investments requiring human clinical trials and FDA approval. He expresses enthusiasm for computational biology but notes that many recent breakthroughs, while computationally aided, weren't solely driven by new computational science 10 years ago. A more pressing issue is the perceived 'brain drain' from the US in scientific talent. Factors like the gutting of research institutions (CDC, NIH), an 'anti-science vibe,' and restrictive immigration policies are pushing top scientists to countries like China. This loss of neurological reserves is detrimental to US innovation, especially as other nations actively recruit global talent.
Venture capital's broken incentives and the future of deep tech
The venture capital industry faces broken incentive structures. Limited partners (like endowments) may prioritize stable, albeit low-return, investments to avoid accountability, while General Partners (GPs) of large funds can earn more from managing a $5 billion fund with a 1.01x return than a smaller fund with a 3x return. This encourages GPs to accept larger checks and higher valuations, even if it’s not optimal for long-term venture returns. For entrepreneurs, accepting massive checks at inflated valuations from large funds can dilute their ownership significantly and potentially create misalignments. Maris believes this pendulum will swing back, and simply focusing on late-stage investing might not be a sustainable long-term strategy. The increased speed and AI enablement are making 'deep tech' areas, like human biology, healthcare, and the underlying AI infrastructure, more tractable for both entrepreneurs and investors.
Mentioned in This Episode
●Products
●Software & Apps
●Companies
●Organizations
●Concepts
●People Referenced
Venture Fund Size vs. Top Decile Performance (DPI)
Data extracted from this episode
| Fund Size | Average Return (4.76x) | Percentage of Top Decile Performers |
|---|---|---|
| Less than $750 Million | 95% | |
| Greater than $1 Billion | 2.42x |
Example Fund Size and Required Exit Value for 3x Return
Data extracted from this episode
| Fund Size | Average Ownership % | Required Exit Value for 3x Return |
|---|---|---|
| $500 Million | 10% | $15 Billion |
| $7 Billion | 10% | $210 Billion |
Common Questions
Bill Maris's new fund is called Section 32. It focuses on being highly selective in investments and people to ensure optimal financial returns.
Topics
Mentioned in this video
Founder of Section 32, former founding CEO of Google Ventures, and former VP of special projects at Google.
Friend who shared photos illustrating the increasing ubiquity of personal cameras.
Co-founder of Android and partner to Bill Maris in conceptualizing Google Ventures.
Co-founder of New Limit, a company in the longevity space.
Co-founder of New Limit, a company in the longevity space, and CEO of Coinbase.
A speaker in the video, referred to as 'Sax'. He discusses venture capital fund strategies and incentives.
Bill Maris's new venture fund, which has raised $150 million and focuses on selective investments for financial return.
The venture capital arm of Google, which Bill Maris founded and led.
A technology news website that published a headline about Google Ventures' innovative approach.
A company invested in by Section 32.
A company in the longevity space invested in by Bill Maris, co-founded by Blake Byers and Brian Armstrong.
U.S. Food and Drug Administration, which requires rigorous safety testing for therapeutic compounds.
U.S. Centers for Disease Control and Prevention, whose funding cuts and an 'anti-science vibe' are seen as detrimental to research.
U.S. National Institutes of Health, whose funding cuts and an 'anti-science vibe' are seen as detrimental to research.
Mentioned in the context of institutional investors who are less likely to face repercussions for investing in large, underperforming funds.
A project incubated by Google under Bill Maris's leadership.
A division of Google dedicated to radical technology. Calico was also incubated here.
A life sciences research and development company associated with Google and founded by Bill Maris. Focuses on longevity.
Mentioned as the place Bill Maris went to buy tar and a mop to fix his apartment roof while operating servers.
A cybersecurity technology company that Section 32 has invested in.
A cryptocurrency exchange platform that Section 32 has invested in.
An agricultural technology company acquired by Monsanto, representing an early successful investment for Google Ventures.
The company that acquired Climate Corp for $1 billion.
An AI research and deployment company whose business model could be threatened by Google's lower token costs.
An AI safety and research company whose business model could be impacted by Google's competitive pricing.
Mentioned as an example of a company that prioritized market share growth over immediate profitability, potentially burning investor cash.
A venture capital firm mentioned for potentially generating extremely high returns on a single fund.
A space exploration company founded by Elon Musk, cited as an example of successful deep tech investment.
A company in the life sciences space that Bill Maris invested in.
A company in the life sciences space that Bill Maris invested in.
Electric vehicle and clean energy company used as an example of a successful deep tech business model.
Mobile operating system co-founded by Rich Miner, who became Bill Maris's partner at Google Ventures.
Google's AI model, which offers significantly lower token costs compared to competitors like OpenAI and Anthropic.
Exchange-Traded Funds that will likely have to invest in companies that are staying private longer.
More from All-In Podcast
View all 419 summaries
32 minPalo Alto Networks CEO: "AI Found 5 Years of Bugs in 6 Weeks"
40 minWhy Secondary Markets Are Eating the IPO | All-In Liquidity Secondary Markets Panel
33 minThe IPO Comeback: Why Tech Giants Are Finally Going Public | All-In Liquidity IPO Panel
32 minDan Loeb: The Lost Art of Short Selling, and Why Stock Picking is Back
Ask anything from this episode.
Save it, chat with it, and connect it to Claude or ChatGPT. Get cited answers from the actual content — and build your own knowledge base of every podcast and video you care about.
Get Started Free