When it comes to artificial intelligence, Silicon Valley’s standard advice for startups might not apply.
Consider OpenAI. Founded in 2015, the A.I. venture behind ChatGPT and GPT-4 is now valued at nearly $30 billion. But its success was hardly inevitable. CEO Sam Altman, who previously led the startup accelerator Y Combinator, discussed his company’s unusual ascent during a fireside chat hosted by fintech company Stripe this week.
“OpenAI went against all of the YC advice,” Altman told Stripe cofounder and fellow billionaire John Collison.
He rattled off the ways: “It took us four and half years to launch a product. We’re going to be the most capital-intensive startup in Silicon Valley history. We were building a technology without any idea of who our customers were going to be or what they were going to use it for.”
Read more: Paul Graham calls A.I. ‘the exact opposite of a solution in search of a problem’
On Saturday, Altman tweeted: “chatgpt has no social features or built-in sharing, you have to sign up before you can use it, no inherent viral loop, etc. seriously questioning the years of advice i gave to startups.”
Asked whether potential OpenAI investors wagged their fingers and told him he was doing it wrong, Altman replied, “Yeah, and I was just sort of like, I don’t really care. Don’t invest.”
Of course, knowing the startup rules in and out allowed Altman to break them with confidence. In addition to leading Y Combinator, whose success depends on evaluating startups, he also served as CEO of Reddit and is a prominent investor—he was an early investor in Stripe.
“Maybe you’re more self-actualized, you don’t have to care so much,” Collison noted, to which Altman replied, “Yeah.”
Greg Brockman, OpenAI president and cofounder, also reflected this week upon the company’s rule-breaking ways.
“You’re supposed to have a problem to solve, not a technology in search of the solution,” he told the Possible podcast this week. He added they spent “a couple months just writing down all the different ideas that we could work on for both GPT-3 and for GPT-4…Maybe we could do a medical thing or a legal thing.”
Instead they decided to ignore the rule altogether—to great success.
A.I. is just different, Brockman concluded: “Every company, every individual, every business is a language business. It has language flows deeply baked in. So if you can add a little bit of value in existing language workflows, then it will just be able to be adopted so broadly.”
