All podcasts / Invest Like the Best / Summary

Brian Chesky - AI Founder Mode

2026-05-05 - 76 min - source - Read full transcript
Patrick O'Shaughnessy (host)Brian Chesky

Key insights

Founders should learn to be CEOs deliberately rather than by trial and error, because trial and error is expensive to unwind.
Chesky argues founders are naturally good founders (an innate skill) but no one is a naturally good CEO. Hiring a professional manager who builds an 'empire' and later has to be unwound can waste years, so the CEO skill set should be learned intentionally rather than discovered on the job.
founder-mode
Founder mode means staying hands-on in the details and giving up control grudgingly, not delegating early and correcting course later.
Chesky says most leaders do the opposite: they let go, hire someone, watch the person go in the wrong direction, and only then intervene, which fails to train the person and wastes their runway. His model, learned partly from how Steve Jobs ran Apple, is to start deeply hands-on and hand off power gradually as trust is earned.
founder-mode
AI founder mode will require even more granular founder attention because near-unlimited execution capacity removes the old excuse for delegating broadly.
Where founder mode meant reviewing every function through recurring in-person group meetings, Chesky expects AI-era organizations to move away from meeting-based culture toward asynchronous work with far fewer management layers, citing the Catholic Church's four layers of management over 2,000 years as a target structure.
founder-mode
Two types of managers will not survive the AI transition: pure people managers with no technical contact with the work, and people unwilling to change.
Chesky argues design leaders like Johnny Ive who both design and lead people are the model; managers who only run one-on-ones and act as therapists to their reports, without engaging the actual work, will be replaced because 'you manage people through the work.'
founder-mode
Consumer AI is underbuilt relative to enterprise AI because of an unclear business model, a false sense that distribution is already mature, and Silicon Valley's trend-following instincts.
Chesky cites a Y Combinator batch of 175 companies where only 16 were consumer-facing. He argues subscriptions face a low ceiling, ads compete with free tools like Claude and Gemini, and e-commerce plays got shut out of third-party app ecosystems, leaving founders defaulting to enterprise because 'that's the trend' rather than because it's the better opportunity.
ai-and-consumer-products
The Eleven-Star Experience exercise pushes an idea to absurdity to make the realistically achievable version look ordinary by comparison.
Starting from a routine five-star Airbnb check-in, Chesky escalates through wine and snacks (six stars), a surfboard and city tour (seven stars), an elephant parade (eight), a Beatles-style airport welcome (nine), up to Elon Musk taking you to space (ten). The exercise works backward from the absurd extreme to find a scalable six- or seven-star version that actually differentiates the product.
product-market-fit
Product-market fit is found by making the addressable problem as small as possible, not by launching at scale.
Chesky says Airbnb, Uber, and DoorDash all started in a single city, and that Airbnb's stalled second and third businesses only took off once he shrank the target from 100 cities to one and applied the resources of a hundred-city launch to a single market. He credits Paul Graham with pushing him toward this small-scale start.
product-market-fit
Project Hawaii used small, dedicated teams and a staged 'crawl, walk, run, fly' process to generate outsized revenue from narrow, well-defined problems.
A roughly ten-to-twelve-person team focused solely on search-to-book conversion generated an estimated $200-300 million in incremental revenue in year one, growing to a run rate contributing roughly 600 basis points of Airbnb's $13-14 billion in gross sales, before the model was repeated on pricing and other problems.
product-market-fit
Hiring is the highest-leverage activity for a CEO, and it should be done through continuous relationship-building rather than reactive search.
Chesky recommends mapping the best people in a field, meeting them informationally, and asking every contact who else is great, building a referral-based pipeline before a role even opens. He also advises sourcing candidates by starting from admired results (an ad you like, a product you admire) and working backward to the person who made it, rather than starting from a resume.
hiring-and-talent
A leader's time spent recruiting is inversely related to time spent managing, because better people are more self-managing.
Chesky says Sam Altman told him early on to spend 50 percent of his time on hiring, advice he initially ignored to his own cost. He now personally co-hires roughly the top 200 people at Airbnb rather than delegating executive hiring, arguing that if executives can't hire people that good on their own, the company hasn't reached far enough.
hiring-and-talent
Chasing adulation is a self-defeating motivational strategy because status-seeking never satisfies, while making things for intrinsic love does.
Chesky describes adulation as 'a cup with a hole in the bottom' that needs ever-larger hits to produce the same high, and recounts feeling empty the day after Airbnb's IPO despite a $100 billion valuation. His stated shift was to stop asking who he wanted to be and start asking what he wanted to make, a reframing he attributes in part to advice from Barack Obama.
creativity-and-motivation

Books referenced

Media referenced

Companies

Techniques and frameworks

Summary

Brian Chesky joins Patrick O'Shaughnessy to trace the arc from his industrial-design training at RISD, through the near-death crisis that forced Airbnb into founder mode during the pandemic, to what he now calls AI founder mode. The core argument is that founders are naturally good at founding but nobody is naturally a good CEO, and that the professional-manager playbook of broad early delegation is actively harmful. Chesky's alternative, absorbed partly from Hiroki Asai (Steve Jobs's longtime but low-profile creative director) and partly from his own near-collapse in 2020, is to stay obsessively hands-on in the details and cede control only gradually, as trust is earned. He argues AI intensifies rather than relaxes this requirement: as execution capacity on-demand grows, organizations will shift from meeting-heavy, hierarchical structures toward flatter, asynchronous ones, and the managers who survive will be the ones with real contact with the underlying work rather than pure people-managers running one-on-ones.

A large stretch of the conversation is about how Airbnb actually builds new things. Chesky's Eleven-Star Experience exercise, escalating a routine five-star check-in through wine and snacks, a surfboard and city tour, an elephant parade, a Beatles-style airport welcome, and finally Elon Musk taking you to space, is a tool for working backward from absurdity to find an achievable six- or seven-star product that genuinely differentiates from competitors. He pairs this with a strong belief that product-market fit only comes from deliberately shrinking the addressable problem: Airbnb, Uber, and DoorDash all launched in a single city, and Airbnb's stalled second and third business lines (services, experiences) only took off once teams stopped trying to launch at hundred-city scale and instead perfected one market first. Project Hawaii operationalized this as a "crawl, walk, run, fly" system: small, dedicated teams (ten to twelve people) applied to narrow problems like search-to-book conversion generated an estimated $200-300 million in incremental revenue in year one, scaling toward roughly 600 basis points of Airbnb's $13-14 billion gross sales run rate.

On talent, Chesky is emphatic that hiring, not managing, is a CEO's highest-leverage job, and that most founders underinvest in it. He describes pipeline recruiting (continuously meeting people and asking each one who else is great, well before a role opens) as superior to reactive search-firm-driven hiring, and recommends sourcing by starting from admired results and working backward to the person responsible, rather than starting from a resume. He personally still co-hires roughly the top 200 people at Airbnb, arguing that if his executives can't independently hire people that strong, the company isn't reaching far enough. He credits (and initially ignored) Sam Altman's early advice to spend half his time on hiring, and frames the tradeoff as direct: time spent recruiting is time not spent managing, because strong hires are largely self-managing.

The episode's most personal thread is Chesky's account of moving away from chasing adulation and toward making things for their own sake. He describes status-seeking as "a cup with a hole in the bottom," something that requires an ever-larger hit to produce the same feeling and never actually satisfies, and recalls feeling strangely empty the day after Airbnb's IPO despite a $100 billion valuation. The shift he made, partly credited to advice from Barack Obama about focusing on what you want to do rather than who you want to be, reframed his motivation around intrinsic craft rather than external validation. He ties this back to bodybuilding in his teens, which taught him that physical self-change is a uniquely high-leverage form of self-change, and that durable progress comes from disciplined, incremental "progressive overload" rather than sporadic intensity.

On the future of consumer AI, Chesky argues the category is structurally underbuilt: a recent Y Combinator batch of 175 companies included only 16 consumer companies, which he attributes to an unclear business model (subscriptions hit a ceiling, ads compete with free tools, e-commerce plays got locked out of app ecosystems), a false assumption that distribution channels are already saturated, and Silicon Valley's tendency to follow trends rather than break from them. He predicts a consumer AI renaissance in the next 12 to 24 months and frames Airbnb's own AI strategy around re-centering the platform on people rather than homes, aiming to build what he calls the most authenticated identity and preference profile on the internet, alongside smaller, faster-moving experimental "sandboxes" outside the constraints of a public, guidance-bound core business.

Notable Quotes

"It's better to have a monopoly of a tiny market than a small share of a big market." - Brian Chesky (citing Peter Thiel)

"Adulation is like a cup with a hole at the bottom. And you keep filling it in, thinking it's love, except it just keeps coming out the bottom." - Brian Chesky

"You manage people through the work. You don't manage the people. Otherwise, what are you doing?" - Brian Chesky

"The two types of people that will not survive the age of AI are pure people managers who think it's all about just leadership, and people that are rigid and don't want to change and evolve." - Brian Chesky

"Don't try to be successful. Try to do something wonderful that you love for yourself. Maybe you'll be successful. Maybe you don't." - Brian Chesky