Venture Capitalist Sarah Guo’s Surprising Bet on Unsexy AI

Detailed Summary

Introduction (0:00 - 1:16)

The hosts introduce Sarah Guo, a rising star in Silicon Valley investing and founder of Conviction, highlighting her background at Goldman Sachs and Greylock. They set the stage for a discussion on AI's future beyond autonomous agents and language models, focusing on Guo's surprising bet on enterprise software.

• Sarah Guo, founder of Conviction, is a prominent Silicon Valley investor.
• She previously worked at Goldman Sachs and the venture capital firm Greylock.
• Conviction, founded in 2022, makes early-stage investments in tech startups.
• Guo believes high-speed change, particularly in AI, is uncomfortable for everyone, including technologists and incumbents.
• She sees software as a service (SaaS) as a prime area for AI-driven transformation.

Founding Conviction and Investment Philosophy (1:17 - 7:11)

Sarah Guo discusses the founding of Conviction, her motivation to return to early-stage investing, and the unexpected success in raising $100 million. She shares insights from her upbringing with tech founder parents, emphasizing the empathy she developed for the entrepreneurial journey and the importance of investors not panicking.

• Guo left Greylock after a decade to pursue focused early-stage investing.
• She initially aimed to raise $50-75 million but secured $100 million, largely from 30 entrepreneurs she had previously worked with.
• Her pitch to investors was a commitment to deliver top-tier venture returns.
• Growing up with tech founder parents instilled in her a deep empathy for the challenges and fears entrepreneurs face.
• A crucial trait she values in partners and investors is the ability to remain calm and not panic during difficult times, as many bad things happen on the journey to building great companies.

The Current Tech Landscape and AI Investment (7:12 - 11:58)

The discussion shifts to the current state of the tech industry, noting the perception of a tech boom driven almost entirely by AI funding, while other sectors remain in a slump. Guo explains Conviction's investment strategy, focusing on the "last mile" of AI implementation rather than competing with massive investments in foundational models.

• The tech industry is in a "panicky time," with significant funding concentrated almost exclusively in AI, while the broader post-pandemic slump persists.
• Guo acknowledges the discomfort of rapid change, including job displacement and competitiveness, but believes embracing technology is the path forward.
• She highlights the massive funding rounds of companies like Anthropic ($3.5 billion) and OpenAI ($40 billion from SoftBank), which are challenging for smaller funds to participate in directly.
• Conviction's strategy avoids direct competition with these large foundational model investments, focusing instead on the "last mile" of AI application.
• Guo believes the "last mile" of bringing intelligence to real-world impact represents 99% of the opportunity, creating numerous companies and value creation opportunities.
• Conviction has invested in early-stage companies like Sierra, Harvey, Cognition, Mistral, and Cartesia, with median entry prices in the millions, not billions.

Judging Untested Businesses and Founder Traits (11:59 - 16:00)

Guo elaborates on her criteria for evaluating early-stage startups, emphasizing the importance of backing talented founders. She stresses the need for founders to have a clear "north star" combined with "paranoia and pace" to adapt to the rapidly changing technological environment, especially in SaaS.

• When choosing between a business idea and founders, Guo prioritizes talent, as the quality of an idea is an expression of the founders themselves.
• Key traits she looks for in founders include a clear "north star" vision and a combination of "paranoia and pace."
• Founders must consistently monitor the technological environment to ensure they remain on track and stay close to their customers and evolving models.
• The current AI era is far less "gentle" than the SaaS investing period 5-10 years ago, where assumptions about software development were less likely to be invalidated overnight.
• She views the AI model landscape as dynamic and competitive, with open-source alternatives like DeepSeek and Mistral challenging established players.

The Future of AI and SaaS Transformation (16:01 - 21:53)

Guo discusses the potential for a diverse ecosystem of AI agents and the transformative impact of AI on enterprise software, particularly SaaS. She uses Salesforce as an example, explaining how AI can automate data entry, customer interactions, and upselling, fundamentally changing current interfaces.

• Guo predicts a diversity of consumer AI experiences, with multiple chatbots for different tasks (work, home, therapy) and many AI features embedded within software via APIs.
• She challenges the idea of a single dominant AI company, noting that no single company has ever commanded all consumer attention in tech history.
• Guo explains that large incumbents struggle to adapt quickly due to the high cost of coordination in vast organizations and the rapid pace of technological change.
• She sees significant, though uneven, impact of AI on SaaS, particularly in areas involving data entry and updates.
• Using Salesforce as an example, she illustrates how AI can automate customer relationship management, populating databases, handling interactions, and even upselling customers automatically.
• Guo believes that if fewer humans are performing data entry and updates, or if these tasks are automated, the fundamental nature of software will change dramatically.

The New Generation of Founders and Capital Markets (21:54 - 29:03)

Guo observes generational shifts among founders, noting a move towards "hard science" and enterprise software, contrasting with the app-based economy of a decade ago. She discusses the deep capital markets for early-stage venture despite a lack of liquidity events and encourages aspiring entrepreneurs to pursue their ambitions.

• Guo observes significant differences in the younger generation of founders, though she humorously questions if it's due to her own aging perspective.
• She highlights companies like Solola, founded by MIT dropouts in their early 20s, focusing on business process automation in "unsexy" sectors like freight brokerage.
• Guo emphasizes that ambition, capability, and strategic thinking are not age-dependent, and she values founders focused on enterprise software and efficiency improvements.
• She notes a trend towards "hard science" businesses, inspired by companies like SpaceX and Anduril, rather than just consumer apps.
• Guo acknowledges the deep capital markets for early-stage venture, with significant inflows into AI companies.
• Despite a lack of liquidity events (IPOs, M&A) in recent years, she remains optimistic about future outcomes, citing Figma and CoreWeave as positive signs.
• She believes that the broader impact of technology and the existence of successful companies like Tesla, OpenAI, and SpaceX will continue to encourage risk-taking among investors and founders.

Political Expression in Silicon Valley (29:04 - 32:00)

Guo addresses the increasing political expression within Silicon Valley, viewing it as a healthy development that fosters diverse opinions and challenges homogeneity.

• Guo observes increased social division and political expression among Silicon Valley individuals and venture capitalists.
• She believes the "Overton window" of acceptable expression has dramatically widened in the last five years.
• Guo views the ability to express true opinions and the resulting less homogeneity in group opinions as a healthy development for the tech industry.
• She notes that some individuals integrate political expression into their business, while others adopt pragmatic strategies, adapting to changing winds.

Conclusion (32:01 - 32:48)

The hosts conclude the interview, and a Salesforce spokesperson provides a clarification on the complexity of their CRM database. A correction is also issued regarding Zipline's autonomous flight miles.

• A Salesforce spokesperson clarified that their CRM database is more complex than a simple list, offering a unified, secure, and customizable customer profile.
• A correction was issued for a previous episode, stating Zipline has flown over 100 million commercial autonomous miles, not 50 million without human pilots.

Notable Quotes

"High-speed change is very uncomfortable, right? Um, it's uncomfortable for human beings. It's uncomfortable for even technologists. It is uncomfortable in particular for incumbents."
— Sarah Guo

"This technical change is going to happen once in my professional lifetime and I can't work on anything else."
— Sarah Guo

"The last mile between like just having some intelligence and having impact in the real world is really long. That's like 99% of it actually."
— Sarah Guo

"You have to have some sort of northstar in terms of what you're building, but also like paranoia and pace in terms of consistently trying to understand what's going on in the environment like the technology environment around you to make sure you are still on track."
— Sarah Guo

"It was a gentler time in the like middle period of SAS investing 5 10 years ago where it was not likely that you would wake up on a Tuesday and be like all my assumptions about how we would build this software were wrong but that is possible today."
— Sarah Guo

"I do not envy the people who have a 100,000 person ship to shift and the technology changes every week."
— Sarah Guo

"The Overton window of like what you could express as a technology person and venture capitalist has opened dramatically over the last 5 years."
— Sarah Guo