Sam Altman, the optimist

No matter how much I plan, I always end up with last minute packing the night before my travels begin. And that's how all sorts of random things find their way into the suitcase. But there is one thing I always manage to plan: reading material. The holidays are the time when we finally, or at least hope to, get through those exciting books we've had "on the list" for the past few months.

And now that everyone is heading off on summer holidays, except for me, 'cause I spent most of it on a trip to Japan in March, it's the perfect time for a book review! But don't worry; we're staying within the tech sphere. This is Kludder, after all. My previous review of the book Careless People, which you can read here (only in Norwegian, sorry!), proved to be popular. So here's a review of the book The Optimist: Sam Altman, OpenAI, and the race to invent the future.

The book first appeared on my radar back in March. The author, Keach Hagey, a journalist at The Wall Street Journal, has interviewed over 250 people in Altman's inner circle, including Sam Altman himself. In the weeks before the book launch, I had a chat with Hagey, who had this to say about the book when we spoke in April.

AI is a global phenomenon, but its current leaders have sprung from the very specific cultural milieu of Silicon Valley. In THE OPTIMIST, I trace how Sam Altman has exemplified this culture since he was a teenager - how he is, as his mentor Peter Thiel put it, 'just at the absolute epicenter' of 'The Silicon Valley zeitgeist. - Keach Hagey to Kludder.

Dropped out

It's not fair to compare The Optimist with Careless People. Sarah Wynn-Williams, who wrote about life inside Facebook, wrote about her own experiences. This gave the book a bite that made it hard to put down, and propelled it straight to the top of bestseller lists.

The Optimist, however, is a more traditional biography. Hagey does a good job of mapping out the people around Altman, from grandparents to ex-boyfriends, classmates, and peripheral investors. For my part, all this "mapping" can be a tad tedious, as Hagey takes considerable time laying out Altman's father's work, and how Altman was seen as a child prodigy from a young age. At the same time, it helps paint a picture and provide an understanding of the man behind OpenAI and the eventually dominant ChatGPT.

The section about Altman's twenties are particularly well written. As a nineteen-year-old, he drops out of his studies to follow the entrepreneur dream. He founded Loopt in 2005, a location service that was supposed to help you find friends nearby. But Loopt never became the success Altman hoped for. In 2012, the company was sold for $43 million.

The student who became CEO

But Loopt gave Altman valuable experience, and it was the "failed" startup that would give him entry into Silicon Valley. Through Loopt, Altman became acquainted at a young age with an investor named Paul Graham. In addition, Altman was taken into the then fledgling Y Combinator, an incubator for startups that advises promising companies on how to succeed. Among the companies that have gone through YC are AirBnB, Stripe, Reddit, and Dropbox. And this is where the story really begins.

Hagey writes in detail about Altman's journey; from starting out as a young entrepreneur, to becoming CEO of Y Combinator and scaling it into the startup powerhouse YC is known for today. These chapters will be enjoyable if you've got an entrepreneurial spark, or just curious about life inside Silicon Valley. It really makes good reading.

The author manages to balance a critical eye on Altman with the inspiring journey he has had. She highlights the results he has created, and the unique ability he has to captivate any investor he meets. For Hagey is not alone in claiming that Sam Altman isn't really an entrepreneur. He's a stellar salesman. Altman's success lies in getting people onboard with his vision. He speaks in a way that oozes confidence.

A good example of this is when he, as a twenty-year-old, sits with yoga-crossed legs in a chair with the venture capital firm Sequoia Capital and enchants them with lofty visions of what Loopt could become. This is how he gets Elon Musk on board in the founding of OpenAI (a friendship that has since gone completely awry). In conversations with Musk, Altman highlights the danger the world faces if Google should win the race for artificial intelligence.

David versus Goliath

Sam Altman makes choices that frustrate others. As leader of YC, his attention is on founding OpenAI. The board at YC feels deceived, and it wasn't solely good wishes that followed Altman out the YC door, in what was essentially – but not quite – a firing.

Altman's choices raises som questions: was it a sort of gray area to use the contact network at YC to get OpenAI off the ground? It's obvious that Sam Altman wants to create something big, do the impossible, and stop Google from winning the AI-race. In hindsight, we can conclude that Google has been shook. For the first time in Google's history, they are experiencing a decline in the number of searches. And that's thanks to ChatGPT, and the advent of language models. Had Altman failed, we might have looked at him as a morally conspicuous leader, someone that let there eyes wander to other projects, someone who's tempted by ever increasing power.

Machiavellian

The Optimist has promised to provide answers to what actually happened in the lead-up to November 17, 2023. Within OpenAI, the event has been called The Blip, where Sam Altman seemingly out of the blue was fired by OpenAI's board. Hagey goes in depth here, and a lot of insight is provided that we haven't had before. Reading it, you're suddenly standing in the middle of the power game that unfolded in the corridors of OpenAI. Board members conspire and look nervously over their shoulders, fearing that anyone could be a spy sent by Altman himself. The whole thing reminds me of the HBO series Succession, known for its many boardroom intrigues and shifting alliances.

To the public, Altman comes across as a true altruist. He has no ownership stake in OpenAI, which is a non-profit research organization. The purpose of OpenAI is to monitor the development of artificial intelligence and ensure a result that benefits all of humanity.

Still, inside the company, Altman is about to achieve complete control of the company he wants to turn into a commercial enterprise.

In the spring of 2023, an OpenAI employee alerts the board that the company doesn't have adequate safety procedures related to language models. One of the measures is to establish a "safety board" with representatives from both OpenAI and partner Microsoft to ensure that the products are not a danger to society. But it turns out that Altman isn't telling the complete truth when he claims that three upcoming products have been approved – it turns out that only one of these products has been submitted and cleared.

Thus we get to know the other side of Altman: The one that is inclined to take shortcuts to gain a competitive advantage. The man who causes Dario Amodei – and a number of others higher up in the OpenAI hierarchy – to leave the company, expressing concern about OpenAI's attention to safety. Today, Amodei is one of the main competitors to Sam Altman, with his company Anthropic and the language model Claude.

The Blip

Mira Murati, chief technology officer at OpenAI, also raised concerns about Altman's leadership. She believes Altman, who is perceived by many as conflict-averse, has a simple playbook he follows: It starts with him saying exactly what you want to hear, to get you on his team. If that doesn't work, he undermines your credibility. By chipping away at the person in question, he hopes to strip away all power and influence the employee, or board member, may enjoy. Murati pointed out that Altman, among other things, promised the same position to two people, which resulted in internal disputes and chaos.

Altman was reinstated as OpenAI CEO on November 22, just five days later. Murati, who said she was willing to take over as CEO after Altman was ousted, left OpenAI not long after his return.

The Optimist gives you a thorough review of Sam Altman's life, and how OpenAI today defines one of our time's greatest technological advances. Keach Hagey brings you right to the center of Silicon Valley, and shows how politics and power struggles play an increasingly larger part of Altman's life. He was the boy who sat in his room tinkering with computers and just wanted to build things.

Now he is the absolute ruler of OpenAI.

Elon Musk's AI nightmare

The Tesla CEO played an important role in getting OpenAI off the ground. He was the one who put in the money, and worked closely with Sam Altman. But then the relationship soured, and Musk withdrew from OpenAI. He bought Twitter, called it X, and created his own language model called Grok.

This week, Grok has spewed out antisemitic content, to the amazement, outrage – and amusement – of X users. According to Wired, updates were made to Grok's code on Sunday. The chatbot has then been instructed to "not be afraid to make claims that are politically incorrect, as long as they are well-reasoned" and to "assume that subjective views obtained from media are biased."

The tigers from China

The New York Times reports from China, Hangzhou, which has become China's own Silicon Valley. Well-known Chinese companies like DeepSeek and Alibaba have headquarters here. They are two of six companies known as "the six tigers from Hangzhou."

The article shows how China is gearing up for the AI race, with Hangzhou as the center of it all. There are plenty of coders there, who call themselves villagers. Daytime is spent coding at cafes, hoping to be snatched up by tech companies. In the evening, they gather for gaming nights.

Several business founders tell The New York Times, anonymously, that the strong support from Chinese authorities scares away investors. It's particularly challenging to get foreign venture firms to invest in them. Because in China you really have two choices:

1. Accept state funding and adapt the product to the Chinese market.
2. Raise enough money on your own to establish offices in a country like Singapore to attract foreign investors.

Who would have thought? Meta wants access to all your photos

Facebook is asking users for access to their camera roll. This way, Meta can automatically suggest AI-edited versions of their images, including photos that you haven't uploaded to Facebook, according to TechCrunch.

By accepting, you consent to Meta's terms for AI use. This means that all your images, with location data, facial features, and other information can be analyzed by Meta's AI tools.