By playing, you agree to uListen's Terms and Privacy Policy and YouTube's Terms of Service.

Sourcery with Molly O'Shea

Elad Gil of Gil Capital, Gil & Co. & Enigma Global

Elad Gil, of Gil Capital and Gil & Co, aka one of Silicon Valley’s most influential investors, sits down with Molly O’Shea to unpack it all..

Elad GilguestMolly O'SheahostMolly O'Sheahost

Oct 17, 202537mWatch on YouTube

Summaries, chapters, transcripts, and FAQs on this page are made by uListen, not YouTube.

Great conversations run long. Their best ideas don't have to.

uListen is free on iPhone, iPad, Android, Chrome, and Claude.

CHAPTERS

Jump to publisher timestamps

Chapter summaries

Technology as a Force for Good and a Career Arc
After a cold open on the 1999 IPO boom, Molly O'Shea asks Elad Gil how he built such a varied career across Google, Twitter, Color Genomics, a book, a podcast and a fund. Gil says he has been around a while and is guided by a view of technology as a net positive for humanity.
- •Molly lists his career stops: Google, Twitter, Color Genomics, a book, early AI investing, a podcast, incubations and a fund
- •Gil says the honest answer is that things accumulate over time
- •He views technology broadly, from agriculture to medicine to running water, as a force for good
- •He has asked where technology can be most useful and let that drive his choices
Building New Public Monuments
Gil describes his project to build monuments in four or five major US cities, arguing that societies at their apex build large public art. He discusses materials, scale, the Eiffel Tower's controversial origins and why placement matters.
- •He says the last large-scale US monument was Mount Rushmore, begun in the 1920s
- •He is experimenting with bronze, marble and carbon fiber, starting at 15 to 20 feet tall
- •The Eiffel Tower was widely considered ugly when built, so polarizing works can become iconic
- •He wants permanent works relevant 200 years from now, placed in major cities
- •He says Bezos's clock is conceptually cool but too isolated
From Angel Investing to Gil Capital
Molly asks how Gil went from angel investing to a large fund. He explains he ran out of his own money, began raising SPVs, and now invests from incubation through pre-IPO, including AI-driven roll-ups, with a small team.
- •He invests from pre-seed or incubation through pre-IPO and has led rounds of half a billion dollars or more
- •He ran out of money as an angel and started raising SPVs, which was tough at first
- •LPs in the SPVs later asked him to do a standalone fund
- •He also funds founders consolidating industries and modernizing them with AI
- •He has four investment people and six or seven on back office and finance, as he is paranoid about compliance and fraud
11:45 – 12:43
Thesis-Driven vs. Opportunistic Investing
Gil lays out two approaches to investing and says he uses both. He cites crypto and generative AI as thesis-driven moves and Airbnb, Stripe and Figma as founder-driven bets, and he describes founders as a distributed search function.
- •Thesis-driven investing makes sense when a sea change in technology appears
- •He began investing in crypto around 2016-2017 and in generative AI around 2021-2022
- •Early AI investments included Perplexity, Harvey, Decagon and Abridge
- •Companies like Airbnb, Stripe, Figma and Instacart came from meeting great founders
- •Thousands of founders act as a distributed search function that one investor cannot beat
Portfolio Size and Investing Across Competitors
Gil estimates he has invested in 150 to 225 companies over 15 years, then explains how he handles competing companies by asking both founders about conflicts. He gives the Square and Stripe example and says conflicts are clearer at later stages.
- •He estimates 10 to 15 investments a year over 15 years
- •He asks both sides whether a potential conflict is acceptable before proceeding
- •About 90% of the time founders expect to compete and don't
- •With Stripe, a Square executive said it was a conflict but Jack said go ahead, and they didn't really compete
- •Later-stage conflicts are clearer, such as Cognition now competing with Coursera
Is Every Software Company Now an AI Company?
Molly asks whether companies must be AI companies to be relevant. Gil says nearly all modern software uses AI to some extent, advises AI founders to take the easy path first, and notes sectors like battery storage where AI matters less.
- •Gil can't think of a modern new software company not using AI
- •Traditional fintech is less active than it once was
- •His advice is to do the dumb easy thing and build defensibility, saving the hard problem for later
- •Defense tech relies heavily on drones and intelligence
- •Base Energy's battery storage is an example of a non-AI-enabled sector
AI's Energy Bottleneck and Global Policy
Asked about bottlenecks in AI, Gil focuses on energy, arguing that solutions like nuclear exist but society chooses not to use them. He says Europe's energy policy hurts its AI training prospects, which may move to the US and the Gulf.
- •Public opinion on nuclear flipped in the mid-1970s
- •The US is 18% nuclear, France 70% and Japan 25%
- •He says deaths traceable to nuclear accidents are very few, fewer than from falling off roofs installing solar
- •Rising European energy costs and German plant closures discourage training data centers
- •Training will likely go where energy is cheapest and densest, such as parts of the US and the Gulf
Government, Regulation and M&A
Molly asks about the administration's approach to technology. Gil discusses regulatory clarity for crypto, state-level AI risks, his limited views on government equity stakes, and his concern that curbs on big-tech M&A hurt startups.
- •Crypto mainly wanted regulatory certainty, which Gil says needs to be codified in legislation
- •State-level AI bills, such as one from Scott Wiener in California, could curtail US AI progress
- •Gil says he hasn't thought deeply about government investments like MP Materials or Intel
- •He thinks curtailing big-tech M&A is negative for the startup ecosystem
Lessons from the 1999 IPO Boom
Gil compares the AI wave to the dot-com era: about 2,000 internet companies went public, yet only a few remain highly relevant. He discusses the backlog of private companies and the idea of perpetually private firms, and how that affects investors and founders.
- •About 450 companies went public in 1999 and another 450 or so in early 2000
- •Only two or three became highly relevant and about a dozen remain important-ish
- •Founders and investors should consider what the AI version of this culling looks like
- •Some firms like Stripe and SpaceX may stay private forever, much like the direct listing wave fizzled
- •Gil has never sold a share of Stripe and says liquidity is available but depends on discounts and pressure
When Founders Should Sell Their Company
Molly raises the HubSpot CEO's tweet about selling now. Gil says founders call him on exits, that advice is contextual, and gives reasons to sell, noting most companies should sell at the right price because most fail.
- •The only good generic startup advice is that there is none
- •People sell because they are tired, because the price is so far ahead, or because of competitive dynamics
- •A handful of companies should never sell, but most should find the right price
- •Traditional venture portfolios saw 50 to 80 percent of companies fail outright
- •Friendster's reported offer of a percent or two of Google is an example of a hard decision
AI Hype, Valuations and Durability
Asked about a 30% Series A premium for AI companies, Gil says new technology waves are simultaneously over- and under-hyped. He cites Azure's AI revenue, uses Braintree versus Stripe to illustrate durability, and says foundation lab valuations look reasonable on revenue multiples.
- •AI is under-hyped given real spending, such as a few billion dollars a quarter on Azure
- •Most startups in this wave will still fail, as in every other wave
- •Gil asks whether an AI company is Braintree or Stripe
- •Short-term success does not guarantee long-term durability
- •On revenue multiples and growth, OpenAI and Anthropic look reasonably valued, though he doesn't know OpenAI's latest numbers
34:29 – 37:52
What It Takes to Build a Trillion-Dollar Company
Gil says a trillion-dollar company comes down to a giant market, then describes founder traits including intelligence, drive, strategic thinking and the ability to read and motivate people. He names Elon Musk as an entrepreneur he admires.
- •The answer to a trillion-dollar company is a giant TAM
- •He sometimes thinks of a founder's life as a Monte Carlo simulation to separate skill from luck
- •Top founders are smart, driven, aggressive, relentless and strategic
- •They read people well and know how to motivate them
- •He admires Elon Musk for rockets, electric cars and co-founding OpenAI

Publisher timestamps

Get more out of YouTube videos.

Read along on your phone, search it beside the video, or ask your favorite AI app what was said.