Podcast Notes | Conversation with Sam Altman: Taking No Equity Stake in OpenAI Was My Biggest Mistake
Compilation & Translation: TechFlow
TechFlow Introduction: As the evolution of Artificial General Intelligence (AGI) and Artificial Superintelligence (ASI) becomes the cutting edge of global technological competition, OpenAI CEO Sam Altman appeared on the popular Indian podcast "Figuring Out," engaging in a nearly 70-minute in-depth conversation with host Raj Shamani. The dialogue not only systematically traces how Scaling Law evolved from an intuitive hypothesis into scientific consensus, but also rarely and candidly addresses public misunderstandings and media controversies surrounding OpenAI's massive bet on compute power, dramatic executive departures, and the social misconceptions sparked by his decision not to hold direct equity in the company.
From betting the company's future during the chaotic period after GPT-4's training but before its public release, frantically hoarding compute resources, to pushing through internal resistance to launch the first-generation ChatGPT with a mere 5% early-test retention rate, Sam Altman deeply deconstructs the underlying mindset behind non-consensus innovation and high-leverage decision-making. Meanwhile, he explores the ambition-cultivation methodology honed during his Y Combinator era, the compounding logic of a personal career, and the psychological and physical toll behind humanity's ultimate endeavor: advancing toward AI-driven scientific breakthroughs (from Navier-Stokes equations to fundamental mathematics).
Guest: Sam Altman (CEO of OpenAI, former President of Y Combinator)
Host: Raj Shamani (Host of a prominent Indian business podcast, entrepreneur)
Podcast Source: Figuring Out with Raj Shamani (https://www.youtube.com/watch?v=TjPPIf-M6KE)
Original Title: Sam Altman On AI Revolution, Superintelligence & Future Of Humanity | FO583 Raj Shamani
Publication Date: November 3, 2026
Key Takeaways
Reflection on relinquishing equity and public misunderstanding: Sam Altman admits that choosing not to hold any equity in OpenAI was initially a practical misstep. Although this intention stemmed from pure mission-driven purpose—viewing it as a pursuit akin to artistic creation—the inherent cognitions of the business world could not comprehend such counterintuitive behavior, instead breeding groundless public suspicion and confusion over hidden motives for power.
The superlinear return law of compute: The most core insight throughout Sam Altman's investment and entrepreneurial career is the astonishing magic of Scale. From the network of companies incubated at YC to early neural networks, anything that functions well at a small scale often yields returns far exceeding linear growth when scaled up dramatically. This is also the first-principle reasoning behind OpenAI's firm bet on large models.
The life-and-death bet post-GPT-4: In the long vacuum period around August 2022, right after GPT-4 finished training but before the world witnessed its capabilities, the team realized the model would open the door to reasoning and Chain of Thought. Amid widespread beliefs that compute scaling was nearing its twilight, OpenAI withstood enormous uncertainty, betting every chip on massive compute clusters.
The counter-intuitive push to launch ChatGPT: The initial beta version of ChatGPT had a retention rate of only 5%, leading management and the research team to generally oppose its release, viewing the product as crude and prematurely timed. Drawing on YC principles, Sam Altman insisted on going live, firmly believing that a product wildly loved by even a tiny fraction must undergo real-world testing and feedback as soon as possible, flaws included.
The trust dilemma of core team departures: Facing the sequential exits and faction-building by top figures like Ilya Sutskever, Mira Murati, Dario Amodei, and Elon Musk, Sam Altman pointed out that this was not merely a conflict of interest, but rather that top-tier minds routinely develop a profound sense of mission paranoia and strategic divergence—"only I can steer it safely enough"—when touching upon technologies like superintelligence capable of reshaping humanity's fate.
From engineering tools to scientific discovery: OpenAI's next most inspiring strategic goal is to make AI an engine for discovering entirely new scientific laws. From breakthroughs in Navier-Stokes fluid dynamics to proofs in frontier mathematical propositions, AI is transitioning from merely processing human-established experience to exploring the fundamental laws of an unknown universe.
Ambition can be learned: One of the most delightful discoveries during my time leading YC is that entrepreneurial drive and ambition can be completely reshaped through environment and community. Placing individuals among peers who dare to tackle absurdly grand goals will utterly shatter their previously narrow cognitive ceilings.
Exponential compounding in careers: Most people simply repeat one year of experience twenty times in their careers, rather than accumulating twenty years of deep compounding. The true rule for success lies in continuously breaking out of comfort zones and using non-consensus decisions at each stage to drive exponential leaps in self-actualization.
The psychological and physical toll of being Sam Altman: Shaping humanity's most critical technology inevitably requires enduring extreme lifestyle pressures and projecting global anxieties. Losing the freedom and leisure of casually working in a café, one must rely on rigorous schedule management and mindfulness meditation practiced in earlier years to maintain a high activation threshold and emotional stability at the eye of the storm.
II. Key Insights
On public perception regarding relinquishing equity: "Looking back, not holding OpenAI equity was a mistake I made. Although my original intention was incredibly pure—like an artist fully immersed in creation, or a top athlete training relentlessly without caring about getting rich—the business world is simply not accustomed to such outliers. When you deviate from the norm, the public will start speculating wildly about what hidden power plays you're plotting."
On trading short-term low status for long-term high status: "The essence of non-consensus innovation lies in this: you must willingly exchange short-term lower social status for long-term, truly elevated standing. When OpenAI was founded, our status was practically negligible. The academic world viewed us as unruly youngsters and claimed AGI was at least a century away, or perhaps impossible. But once reality proves you right, being misunderstood by the entire world becomes your strongest moat."
On the courage to launch the first ChatGPT: "If you don't feel slightly embarrassed by the first version of a product you ship, it means you launched too late. At the time, ChatGPT's beta retention was brutally low at 5%. Every data point screamed not to launch. But I firmly believed in YC's golden rule: It's better to have five thousand people hopelessly fall madly in love with you, rather than one million people mildly like you."
On the Shakespearean drama behind superintelligence: "The journey to build superintelligence cannot be devoid of maddening twists and turns. When the stakes rise to jeopardize human civilization's future, many top-tier brains develop a psychological pattern: they are unwilling to hand over the steering wheel to anyone else but themselves. This mistrust of concentrated power and technological roadmaps has spawned the industry's epic waves of alliances and conflicts."
On the ultimate pursuit of AI exploring unknown sciences: "What gets me most excited isn't just building a chat assistant, but enabling machines to help us understand the universe. The astonishing capabilities AI demonstrates on challenges like mathematics and the Navier-Stokes equations indicate that we are much closer to the singularity where machines autonomously discover new physics insights, crack complex diseases, and unlock clean energy, than anyone anticipated."
On the compounding rule for breaking cognitive mediocrity: "The tragedy for many is that they claim twenty years of work experience, but in reality, they've mechanically repeated the first year's experience twenty times. You must force yourself to stay continuously on the exponential growth curve, even if every stage brings immense discomfort and complete novelty."
III. Transcript
1. Reflections on Misunderstandings Relinquishing Equity: Why Is Selfless Contribution Viewed as Anomalous in the Business World?
Raj Shamani: When reading coverage about you in the US and India, I noticed a highly polarized phenomenon: You sparked the greatest technological revolution of this era, yet didn't even take a single share of the company you founded. Somehow, however, many people suddenly started framing you as a villain. How did this happen? Why do you find yourself in this position?
Sam Altman: I believe this stems largely from American society's natural wariness of power, particularly the high degree of skepticism directed at the vast control wielded by today's tech giants. On one hand, maintaining such oversight and vigilance is a good thing, but sometimes it does feel like it crosses the line.
Raj Shamani: So what do you think is the biggest misconception outsiders have about you?
Sam Altman: Frankly, what you just mentioned is indeed one of the biggest misunderstandings. Looking back now, deciding not to hold direct equity in OpenAI was actually a pragmatic misstep. Back then, I believed I had profoundly noble and sufficient reasons for this choice, but it ended up bringing tremendous confusion and chaos to the outside world.
Raj Shamani: Why did everyone misunderstand this decision? What exactly couldn't they grasp?
Sam Altman: Because it completely deviates from mainstream thinking patterns. In traditional perception, a person founding a company and grinding tirelessly as a business leader has only one rational motivation: acquiring massive financial rewards. This behavioral pattern clashes too violently with conventional business logic, leaving the public inherently uneasy about incomprehensible outliers. If an artist pours their heart into pure art, or a sports star trains day and night for competitive glory without chasing wealth, everyone accepts and respects that. But in business, people start suspecting: If he's not after money, what is he after? Is he secretly scheming for greater power?
Raj Shamani: Then what is your true driving force?
Sam Altman: Because it's the most fascinating, significant, and socially beneficial undertaking I can imagine. To personally participate in and drive such an exploratory journey is in itself unparalleled thrill and luck. For me, it's the greatest feat worth dedicating a lifetime to.
Raj Shamani: You famously wrote some time ago: You should be willing to trade short-term low status for long-term elevated standing. As long as you're right, being misunderstood by the masses is a huge advantage rather than a disadvantage. But the key question is: Before the final outcome is known, how do you know you're just being misunderstood rather than completely wrong?
Sam Altman: The truth is: until everything settles, you can never be 100% certain. You can only strengthen your conviction, incrementally increasing certainty as you gather new evidence. But the risk of the unknown is precisely what makes this adventure most thrilling. When we first founded OpenAI, this was entirely a low-status endeavor. Everyone treated us as a joke, claiming AGI was at least a hundred years away, viewing us as just a bunch of uncredentialed kids daydreaming. But we saw physical laws others missed. As we solved more puzzles, our confidence grew. Even though the probability of ultimate success seemed infinitesimal back then, its expected value was so high that it warranted betting everything.
2. Superlinear Returns of Compute: From YC Experience to Intuition Breaking Through Scaling Law
Raj Shamani: While the whole world was pessimistic about AGI, what underlying signals made you so confident it would definitely work, willing to bet your entire career on it?
Sam Altman: I believe a person might only ever gain a handful of crucial, great insights in their lifetime, and what you can do is apply them repeatedly across different dimensions. For me, the most central one is this: Scale possesses incredible superlinear emergent properties.
Raj Shamani: Where did this insight originally come from?
Sam Altman: I have personally witnessed this law across countless domains, including corporate growth, national rise, urban evolution, personal careers, and the investor-alumni network I built at Y Combinator. Any mechanism that functions well at a small scale often bursts forth with staggering excess returns once you scale it up several or dozens of times. As someone fundamentally wired for pattern recognition, when we early on saw neural networks not only functioning but growing smarter simply by expanding parameters and data, I developed an extremely strong intuition: We must push scale to its absolute limit at all costs. Though we hadn't yet derived rigorous mathematical formulas for Scaling Law, the simple rule that making networks bigger leads to capability leaps already showed cracks back in the AlexNet era.
Raj Shamani: But surely top scholars worldwide could see this signal. Why didn't other major tech companies and institutions push forward aggressively at the time?
Sam Altman: This is one of the most baffling phenomena of my life. Going back to 2015, Google internally almost possessed the world's top deep learning talent and computing reserves. Yet, the traditional academic circle held a deeply rooted disdain for simply brute-forcing compute. They believed true intelligence must rely on elegant new algorithms, symbolic logic, or bionic structures, viewing a thousand-fold hard increase in compute as technically barren heavy lifting. This intellectual arrogance gave us a precious four-to-five-year window in an undisputed wilderness.
3. The Life-and-Death Bet After GPT-4: Betting Everything in a Vacuum Period Lacking Understanding
Raj Shamani: Could you recreate one of the most critical life-or-death moments for us? Faced with this decision, the outside world warned you absolute failure. Yet, relying on half certainty and half perceived madness, you made the move that completely altered OpenAI's destiny?
Sam Altman: The most heart-stopping decision occurred right around August 2022, just as GPT-4's training wrapped up. At the time, we were still roughly seven to eight months away from officially releasing GPT-4 to the world. Internally, we were stress-testing the model, and everyone knew the intelligence it displayed was unprecedented. More critically, it took us a while to profoundly realize the entirely new research dimension it opened: Models would continue evolving post-GPT-4 at unimaginable speeds. We could absolutely leverage it to develop Chain of Thought and entirely new architectures equipped with deep reasoning capabilities.
Raj Shamani: What kind of strategic choice did this require of you back then?
Sam Altman: We decided to bet the entire company's future on an unimaginably massive procurement of compute power. At that moment, the outside world hadn't even seen GPT-4, let alone understood its capabilities. Mainstream discourse was still claiming large language models had hit a wall. Everyone thought we were insane, spending billions of dollars in advance to order unproven ultra-scale compute clusters, essentially suicide. But because we had intimately used GPT-4 and fully internalized the certainty of Scaling Law, we concluded we were likely right.
Raj Shamani: If you had backed down then, what was the alternative?
Sam Altman: The alternative was to make a very conservative, safe, conventional compute investment and slowly iterate the models. If we had chosen that path, OpenAI would never have reached the cutting edge of AGI today. Often, the true watershed moment is whether you dare, during the long vacuum period before market awareness awakens, to unhesitatingly push all your earned chips to the center of the table.
4. The Counter-Intuitive Decision to Push ChatGPT: The Disruptive Signal Behind 5% Retention
Raj Shamani: Was there any core decision that the entire leadership and research team opposed, yet you forcefully pushed through against all odds?
Sam Altman: Launching the first generation of ChatGPT itself is the most typical example. It wasn't that the team didn't want to build a conversational bot at all, but rather that the vast majority felt it was too crude back then, and the technology wasn't mature enough. As a serious frontier research institution, releasing such a rough demo to the public was considered beneath our standards and more likely to distract focus from core AGI research.
Raj Shamani: What exactly drove you to defy all objections and insist on launching? Was it your hands-on experience from YC?
Sam Altman: Exactly, it was purely the YC ethos at play. During those years at YC, I instilled in founders countless times: If you don't feel the slightest embarrassment about the first version of a product you ship, it means you're launching too late. Because working in isolation is fatal; what you need most urgently is real-time feedback from actual users. When we internally tested the ChatGPT prototype, the metrics were abysmal, with week-two retention sitting at a dismal 5%. Any conventional product manager seeing that number would call it dead instantly, declaring the product unlaunchable.
Raj Shamani: With only 5% retention, why did you see hope instead?
Sam Altman: Because YC holds another battle-tested iron rule: Early in a product's life, having a tiny handful of core users who go crazy for you is vastly more valuable than having a mass of lukewarm admirers. When we followed up with that 5% retaining cohort, we found they were virtually living inside the app, unable to leave. This intense usage depth was an extremely strong counter-intuitive signal. So I made the call to publicly release it as a Research Preview. History proved that this iterative deployment strategy, accepting real-world forging as early as possible, not only ignited the global AI wave but also provided us with irreplaceable safety-evolution data.
Raj Shamani: Now that OpenAI is so massive, would you still be willing to launch products that make you feel awkward today?
Sam Altman: For the core ChatGPT version boasting hundreds of millions of daily users and enterprise-grade critical infrastructure, the bar is already extremely strict; absolutely zero room for sloppiness. But regarding peripheral experimental products, I still strongly advocate agile exploration. In the future, you will still see us constantly rolling out bold, even immature new attempts. Some will be stunning and integrated into core experiences, while others may declare failure. If a company completely loses the courage to ship rough products out of fear of making mistakes, it is not far from mediocrity and obsolescence.
5. The Shakespearean Drama of Core Team Departures: Trust and Power Dynamics Under Superintelligence
Raj Shamani: OpenAI once gathered some of the smartest geniuses in the world, yet subsequently went through a wave of executive departures dubbed a "Shakespearean drama" by the media. Ilya Sutskever left, Mira Murati left, Dario Amodei founded Anthropic, and Elon Musk struck out to build xAI. Initially united by a shared lofty ideal, they are now fighting independently or even suing each other. What is the core divergence behind this?
Sam Altman: I think when a technology's prospect becomes so vast and the stakes so high, touching the future fate of human civilization, a very specific psychological mechanism triggers in the subconscious of many geniuses: They begin to trust only themselves, firmly believing that only by personally steering it can they ensure this technology goes straight.
Raj Shamani: Does this psychology stem from disagreements on technical safety roadmaps, or fear of excessive power concentration?
Sam Altman: The two are actually deeply intertwined. Psychological projection usually runs like this: Because I perceive flaws in his technical philosophy or pace of advancement, I absolutely cannot allow such immense superintelligence control to concentrate in his hands. Differences in roadmaps naturally spawn a collapse of trust.
Raj Shamani: Having weathered boardroom coups, litigation storms, core ally defections, and scrutiny from global governments, how has your own mindset transformed? You seem incredibly composed regardless of the vortex size. Is this just your innate personality?
Sam Altman: No one can live a calm life in the process of building superintelligence; chaos and upheaval are inevitable byproducts of this task. When I first founded OpenAI, I wasn't emotionally prepared. But once Scaling Law was validated and the mission became irreversible, I knew clearly my life would inevitably be swept into hurricane-force winds. This is indeed brutal, but my nervous system seems innately adapted to high-intensity chaotic environments. No matter how the winds rage around me, I can maintain core clarity.
Raj Shamani: Do you really not feel tense ordinarily? When was the last time your heart raced?
Sam Altman: Of course, but the threshold for making my heart race has become extremely high now. The last time my nerves truly tightened was on the eve of preparing to officially announce our large model's breakthrough in Navier-Stokes fluid dynamics. That is a foundational physics problem humanity hasn't thoroughly cracked in centuries. Seeing our model truly step into the deep waters of human scientific research, that historic witness thrill indeed made my heart race in quite some time. Years ago, I spent two hours daily on deep meditation. This long-term focus training allows me, even under extreme pressure, to quickly close my eyes within seconds and enter flow state, completely shielding emotional noise.
6. AI's Next Milestone: From Assisting Engineering to Discovering Entirely New Scientific Laws
Raj Shamani: If we step beyond current commercial competition, within the complete landscape of AGI and superintelligence, what is the next extraordinary thing you personally most desire to create with your own hands?
Sam Altman: My most hauntingly desired goal is to explore exactly how far we can push AI into unknown frontiers, to let it autonomously discover entirely new objective scientific laws. The early results we've seen in frontier mathematics and theoretical physics are profoundly impactful, firmly convincing me that we are much closer to this milestone than the outside world imagines.
Raj Shamani: Does this mean AI will completely transcend its attribute as an engineering tool?
Sam Altman: Exactly. Every leap in quality of life and prosperity humanity has achieved over the past few centuries has been fundamentally driven by breakthroughs in basic science. If we can build a superintelligence machine that not only summarizes human existing knowledge but also helps us penetrate deeper physical mysteries of the universe, invent revolutionary medical protocols, and conquer infinite clean energy, it will completely rewrite the historical trajectory of human civilization. To personally participate in building this machine that explores universal secrets is my most sacred mission.
7. Ambition Can Be Taught: How YC Reshapes Entrepreneurs' Cognitive Limits
Raj Shamani: You've mentioned that the primary reason most entrepreneurs fail is a lack of drive and ambition. But isn't ambition innate? Can a person really be cultivated to become more ambitious later in life?
Sam Altman: The answer is an absolute yes. This is also one of the most delightful discoveries from my years working at Y Combinator. The reason many appear unambitious is that the environments, schools, and social rules they grow up in constantly suppress their nature, forcing them to stay within the flock, treating mediocrity as a legitimate safe haven.
Raj Shamani: What exactly is the magic YC uses to change a person's ambition dimension in just a few months?
Sam Altman: Environmental disruption. When you throw a young person who merely wanted to build ordinary software into a cohort where everyone around is seriously discussing how to cure aging, colonize Mars, or build new global financial infrastructure, their cognitive boundaries regarding possibility are shattered in an instant. They literally see with their own eyes that those achieving earth-shattering accomplishments aren't mythical beings with extra arms, but flesh-and-blood ordinary humans. Ambition is highly contagious. Once inner fears are broken and people realize attempting seemingly absurd grand goals isn't statistically terrifying, a person's inner potential will release explosively.
8. Career Compounding and the Price of Fame: The Psychological Reconstruction of Becoming Sam Altman
Raj Shamani: You wrote a widely circulated blog post early on titled "How to Get Successful," mentioning 13 rules. If you had to distill it down to the core one today, which would it be?
Sam Altman: Without a doubt, the exponential compounding of personal growth. Vast majority severely underestimate compounding's power. They fail to urge themselves to continuously push forward, eventually hitting a dead end in career development. The cruelest irony in the workplace is: Many boast twenty years of senior experience, but in reality, they merely mechanically repeated their first year's experience twenty times.
Raj Shamani: How can one ensure they always stay on the compounding curve, rather than stagnating?
Sam Altman: The most critical factor is daring to force yourself to embrace brand-new, highly uncertain challenges at the end of each stage. Almost everyone, once tasting a little sweetness, immediately hides in a comfortable harbor and stops evolving. You must continuously do things that make you uncomfortable and defy consensus, using the momentum accumulated from the previous stage as a lever to jump to the next. Only then will your cognition and influence exhibit an exponentially growing slope.
Raj Shamani: As someone potentially shaping humanity's most critical technology, what is the true personal cost of becoming today's Sam Altman?
Sam Altman: Subjectively, this is simply my normal life now. Those relaxed, leisurely days from early years have turned into distant, blurry memories. The minutest cost, for instance, is that I can no longer quietly work solo in a corner café on weekends, encased entirely by minute-by-minute security and schedules. The heavier cost lies in the fact that fear, anxiety, and projection regarding the unknown from all global strata crash over me like a tidal wave almost daily. It is genuinely exhausting and heavily pressing, but I deeply know I possess the world's most incredible opportunities and missions. I never blame heaven or others, nor do I feel I endure immense suffering. Since choosing to stand on this historical stage, my only path is to run toward the future with full concentration.