
[尾原オススメ他番組]【ChatGPT作った男】サム・アルトマンの半生と伝説エピソード | OpenAI創業者
- Overview This episode of ハイパー起業ラジオ (Hyper Entrepreneurship Radio) presents a deep div...
- The central thesis is that Altman was not merely a lucky or technically brilliant fou...
- The conversation, hosted by IT critic Obara Kazuhiro and sub-host Kensuu, has the fee...
Readers who want the substance of a podcast episode before listening.
ハイパー起業ラジオ / 尾原和啓 / けんすう
Overview
This episode of *ハイパー起業ラジオ* (Hyper Entrepreneurship Radio) presents a deep dive into the early life and career of Sam Altman, the CEO of OpenAI, based on the book *The Optimist* (written by someone whose Japanese translation had not yet been released at the time of recording). The central thesis is that Altman was not merely a lucky or technically brilliant founder, but a figure whose outsized ambition, strategic networking, and almost preternatural ability to attract powerful mentors—from Paul Graham to Peter Thiel—set him apart even before he achieved major success. The conversation, hosted by IT critic Obara Kazuhiro and sub-host Kensuu, has the feel of two fascinated tech insiders trading anecdotes and marveling at Altman's trajectory, blending admiration with critical observations about his personality and management style.
Early Childhood and Prodigy Signs
The episode opens with a detailed look at Sam Altman's childhood, painting him as an unmistakable prodigy. Born in 1985, Altman is a millennial, just one year younger than Mark Zuckerberg. His parents described him as no ordinary child. At just two years old, he was reportedly able to insert a videotape into a VCR and play it by himself—a feat the hosts jokingly compare to someone "on their second playthrough of life." By age three, Altman had grasped the concept of area codes, asking his parents which number to dial before they even made a phone call. The hosts note that most people, even the quickest learners, do not understand area codes until elementary school.
A third anecdote is even more striking: his parents claimed that if they had dropped him off in New York City at age ten, he would have completely understood the city's structure and never gotten lost. The hosts compare this to navigating Shinjuku Station in Tokyo, a labyrinth that confuses even university students. From a very young age, Altman demonstrated an ability to scan and fully comprehend complex systems.
Competitive Drive and Early Tech Fascination
Altman's competitive nature was evident from childhood. He was described as a boy obsessed with winning, telling his siblings during board games and card games, "I will win, I will run everything." The hosts frame this as a manifestation of a "king's disposition"—an innate desire for control and victory. This trait would later be echoed by Paul Graham, who famously said that when it comes to design, you ask "What would Steve Jobs do?" but when it comes to strategy and ambition, you ask "What would Sam do?"
His engagement with technology was equally precocious. At age eight, he received a Mac and taught himself BASIC, a programming language. Bored with its simplicity, he began to imagine a future where computers would think for themselves and code would no longer need to be written line by line—a vision of AI that he held even as an elementary school student. By age 14, he had moved on to the more complex C programming language and was already debating AI with his teachers.
University, Life's Mission, and the First Startup
Altman was accepted to both Harvard and Stanford, the two most elite universities in the United States. He chose Stanford. While there, he wrote a list of the topics he wanted to dedicate his life to. The list, in order, was: AI, nuclear energy, and education. The hosts note that this list appears two or three times in the book, with slight variations—sometimes including hair loss treatment, which they found amusing—but AI was always at the top.
He began working under Andrew Ng, the future founder of Google Brain, on a project involving autonomous helicopters. This project led him to become fascinated with GPS and location data. Around 2005, when mobile phones began to include GPS functionality, Altman realized this new capability could be the foundation for a service. He brainstormed with a college friend about what the most useful application of mobile location data could be. They ruled out building a Google Maps competitor as too difficult and eventually settled on the idea of a social network where friends could share their locations. This was the prototype of what would later become Loopt.
Y Combinator and the Paul Graham "3-Minute" Impression
At age 19, Altman applied to the very first batch of Y Combinator (YC), which would later become the world's number-one startup accelerator. He was invited for an interview, and the founders of YC were instantly captivated. The legendary episode involving Paul Graham occurred here: within three minutes of meeting Altman, Graham reportedly said, "This must be what Bill Gates was like at 19." The hosts are astonished by this—Graham, a highly respected figure in the programming and startup world, made this judgment in just 180 seconds.
Altman's co-founder at the time described him as having "challenging confidence and a first-class bullshit filter." The hosts speculate that his confidence was so extreme it bordered on a kind of "chaos energy." This confidence, combined with his undeniable presence, got him accepted into YC. Loopt was born. The other startups in that first YC batch included Reddit (founded by Alexis Ohanian) and Twitch (founded by Justin Kan). Even among this legendary cohort, Altman stood out. While the others were tech-loving geeks immersed in programming, Altman was described as more business-minded and success-driven than anyone else.
The Sprint Deal: A Masterclass in Sales and Persistence
The hosts share a remarkable story that illustrates Altman's business acumen. In the pre-smartphone era, Loopt needed to partner with mobile carriers to access location data via APIs. Altman, then 19 and small in stature (looking about 12 years old), was introduced to the CTO of Sprint, the third-largest US carrier. The CTO initially asked, "Where is the founder?" assuming Altman was the founder's son. But once Altman started talking, pitching Loopt as "a mobile version of Facebook with a twist," the CTO became completely captivated and agreed to give it a try.
However, when Altman contacted the executive at Boost Mobile (a Sprint subsidiary) to formalize the partnership, he was told they were on the verge of signing with a competitor. The competitor could not offer one feature: a notification when friends are nearby. Altman's team worked through the night to build that exact feature. The next day, they flew to Boost's office without an appointment, showed up unannounced, and asked for just ten minutes. Altman sat cross-legged on the floor of the conference room—a posture described as "Indian style" or "agura"—and began to pitch. The Boost employees later recalled that he "radiated confidence." By the end of the meeting, the executive told his boss that Boost had to change its strategy. Altman won the contract. The hosts compare this relentless persistence to Jensen Huang of NVIDIA, who famously refused to give up on winning Microsoft's business.
The deal was massive: for the next four years, Loopt would be pre-installed on every Boost mobile phone. Boost also ran a large-scale TV advertising campaign, and within three months, Loopt acquired over 100,000 users—a huge number in the pre-iPhone era when there was no App Store.
The Steve Jobs Encounter: From "Weak" to "Cool"
Despite the early success, Loopt faced challenges. Privacy concerns led users to turn off location sharing, and the service was only available on Sprint phones, far from achieving product-market fit. Then, in 2007, the iPhone was announced. A team member suggested building an iPhone app, and Altman agreed. He also began seeking a partnership with Apple.
Through Sequoia Capital (a legendary venture capital firm that had invested in Apple and was an investor in Loopt), Altman was introduced to Steve Jobs. Jobs, who disliked social networks, initially dismissed Loopt with a single word: "It was weak." However, some younger Apple team members were interested and invited Altman to discuss what the iPhone SDK could do. To get Jobs's official permission to present at Apple's WWDC event, Altman had to pitch directly to Jobs again. He was so nervous his mouth went dry, but he delivered a thoroughly rehearsed pitch. Jobs paused and said one word: "Cool."
The hosts revel in this moment—the upgrade from "weak" to "cool" from Steve Jobs. Altman, then 22, was allowed to present at WWDC. To make himself memorable, he wore two polo shirts layered on top of each other—one bright pink, one neon green—a fashion choice that has since become a meme. It worked: people remembered "that weird guy," and he became a star. Apple also paid for national TV commercials for Loopt, and its downloads surged, briefly outpacing even MySpace. For a moment, Loopt was a king of social networking.
But Jobs was demanding. He required Loopt to work in every language and every country, a technically difficult task at the time. When Altman failed to meet this requirement, Jobs reportedly threw a pen at him in anger. Altman later said that his meetings with Jobs were the most difficult he had ever experienced. Even the famously confident Altman was shaken and trembling. Yet the hosts argue that this three-month period of working with Jobs had a profound impact on Altman, allowing him to absorb Jobs's intensity and learn how to wield power.
The Downfall of Loopt and the First Coup Attempt
Despite the Apple boost, Loopt's fortunes turned. The Lehman Shock cooled the fundraising environment. Facebook offered to acquire Loopt for 15 billion yen (approximately $150 million at the time). Sequoia's Michael Moritz asked Altman what he thought. Altman replied, "Pass." The hosts note that this is a legendary moment: even a legendary investor like Moritz was surprised, as founders are usually eager to sell. But Altman, driven by a fierce will to win and a desire to build an independent giant, refused to sell to Zuckerberg's Facebook.
However, Loopt's momentum was fading. Competitors like Foursquare were rising. Altman himself became distracted, developing a game-like dating app on the side. Colleagues began calling him "Kirakira Shokogun" (Glitter Syndrome)—a term for someone who gets swept up in fame and starts chasing too many things. He was more interested in hitting home runs than generating short-term revenue, and employees grew worried.
This led to a boardroom coup attempt. Employees and investors tried to remove Altman as CEO. The hosts draw a direct parallel to the later OpenAI coup. At Loopt, the reasons were similar: Altman was accused of saying things that were not true, being harsh with employees, and failing to focus on the core business. Sequoia, however, was reluctant to fire him for two reasons: first, Altman was Paul Graham's favorite at YC, and antagonizing him could damage Sequoia's relationship with YC; second, Altman was serving as a valuable scout for Sequoia, constantly meeting with startups and referring promising deals. His networking value outweighed his shortcomings as CEO. The coup fizzled.
The Stripe Investment: 150 Million Yen Becomes 100 Billion Yen
During the Loopt era, Paul Graham introduced Altman to Patrick Collison, the future co-founder of Stripe. Altman impressed Collison by demonstrating deep knowledge of Lisp, a programming language favored by tech enthusiasts. Although Altman was not a technical expert, he had studied enough to seem remarkably knowledgeable. Later, at Paul Graham's house, Collison and Graham were discussing the idea for Stripe. Altman happened to join them, and Graham said, "Sam, why don't you invest too?" Altman put in about 1.5 million yen (roughly $150,000 at the time). That investment in Stripe, one of the most successful startups of the last two decades, eventually grew to be worth over 100 billion yen (approximately $1 billion). The hosts are stunned by the sheer luck and timing: "You just go to a friend's house, an uncle says 'put some money in,' and it turns into 100 billion yen? What kind of world line is that?"
Altman also helped Stripe with fundraising. Collison later said that while he treated Peter Thiel and Michael Moritz like gods, Altman never flinched in front of them and gave useful advice. The hosts note that the only person Altman ever seemed intimidated by was Steve Jobs. Against everyone else—Thiel, Moritz, Musk—he was utterly fearless.
The End of Loopt and the Rise of a New Patron
Loopt was eventually sold to Green Dot, a publicly traded company, for about 4 billion yen ($40 million) in an "acqui-hire" deal. Altman personally netted around 500 million yen ($5 million). By the standards of Silicon Valley legends, this was a modest outcome. Sequoia offered him a job as a venture capitalist, but he declined. Instead, he used his connection with Peter Thiel—whom he had bonded with over discussions about nuclear fusion—to start his own personal fund.
Thiel's praise for Altman was extraordinary. Thiel, the early investor in Facebook, said that if you had to name one representative figure of the millennial tech generation, it would be Sam Altman, not Mark Zuckerberg. He called Altman the "center of the Silicon Valley zeitgeist." The hosts are baffled: Zuckerberg had built Facebook, a global empire, while Altman had exited a modest startup for $40 million. Yet Thiel, a man of immense influence, ranked Altman higher. This, they argue, is the central mystery of Altman's career: how did he attract such powerful patrons—first Paul Graham, then Peter Thiel—before he had any major success?
Y Combinator President and the Pivot to Deep Tech
With Thiel's backing, Altman built a personal fund and invested heavily in YC alumni, earning it the nickname "YC's index fund." It was highly profitable. In 2014, Paul Graham, who had been grooming Altman, named him the next president of Y Combinator. Many YC participants were confused: "Who is this guy?" Altman himself was hesitant—he wanted to start a nuclear energy company and did not see himself as an investor. But influenced by Peter Thiel's argument about "technological slowdown" (the idea that innovation was stagnating), he decided to take the role and expand YC's focus.
Under Altman, YC shifted from primarily funding software companies (like Stripe and Airbnb) to investing in "deep tech" and "hard tech." His investment priority list placed nuclear energy first and AI second. This pivot was partly inspired by DeepMind, which had been acquired by Google. Altman had discussed DeepMind with Thiel, who had been an early investor, and realized the immense potential of AI. He wrote an essay at the time stating that "not enough smart people are working on AI, given its potential impact." The hosts emphasize that Altman recognized the opportunity in AI very early, thanks to his privileged position in the Thiel-Musk-Hassabis network.
Conclusion
This episode matters because it reframes Sam Altman not as a sudden success story of the ChatGPT era, but as a figure whose entire life was a preparation for a grand stage. The key takeaway is the power of strategic networking and the ability to attract mentorship from titans like Paul Graham and Peter Thiel before achieving any major success. Altman's story is also a cautionary tale about the dangers of distraction and the recurring pattern of boardroom conflict—a pattern that would repeat dramatically at OpenAI. The hosts leave the listener with a sense of awe at Altman's trajectory, but also with a clear-eyed view of his flaws: his tendency to say things that are not true, his difficulty focusing, and his relentless, sometimes ruthless, drive to win.
要点
- Sam Altman was a prodigy from age two, demonstrating advanced cognitive abilities like understanding area codes at three and predicting AI at eight.
- His first startup, Loopt, was a location-based social network that failed to achieve massive success but taught him critical lessons about sales, persistence, and the dangers of distraction.
- Altman's ability to attract powerful mentors—Paul Graham (who compared him to Bill Gates after three minutes) and Peter Thiel (who ranked him above Mark Zuckerberg)—was a defining factor in his career.
- His investment of ~$150,000 in Stripe, made almost casually at Paul Graham's house, grew to be worth over $1 billion, providing the financial foundation for his later ventures.
- Altman's encounter with Steve Jobs was a transformative experience: he went from being dismissed as "weak" to being called "cool," but also learned the intensity of dealing with a demanding genius.
- The pattern of boardroom coups and accusations of dishonesty that later defined the OpenAI drama first appeared at Loopt, where employees tried to remove him as CEO.
- As president of Y Combinator, Altman pivoted the accelerator from software to deep tech, prioritizing nuclear energy and AI, setting the stage for OpenAI's creation.