Sam Altman says the OpenAI board shrank from nine members to six over about a year because the members couldn't agree on additions.
“It was nine and then it went down to six and then we couldn't agree on who to add.”
Person
Sam Altman
2 episodes · 40 quoted claims
Sam Altman says Larry Summers was suggested for the new board by the old board members during the crisis weekend.
“Larry was actually one of their suggestions, the old board members”
Sam Altman says two board members called him on the Saturday after his firing to discuss his returning to OpenAI.
“Saturday morning, uh, two of the board members called and said, Hey, we, you know, destabilized.”
Sam Altman says OpenAI's unusual structure evolved by repeatedly patching a nonprofit as it needed far more capital.
“How do we patch the structure? Um, and then patch it again and patch it again.”
Sam Altman says Elon Musk wanted to make OpenAI a for-profit company under his control or merge it with Tesla.
“He wanted to make OpenAI into a for profit company that he could have control of or have it merge with Tesla.”
Sam Altman says Grok, from Musk's xAI, had not been open sourced until people pointed out the hypocrisy.
“Grok had not open sourced anything until people pointed out it was a little bit hypocritical”
Sam Altman said OpenAI uses lots of human-labeled data in its work, without giving specifics about how Sora is trained.
“I mean, without saying anything specific about the Sora approach, we we use lots of human data in our work.”
Sam Altman argues the right way to measure AI's economic impact is the share of tasks it can do, rather than the share of jobs replaced.
“The way I think about it is not what percent of jobs AI will do, but what percent of tasks will AI do?”
Sam Altman says most users don't need GPT-4 Turbo's 128K-token context window most of the time.
“Most people don't need all the way to 128 most of the time.”
Sam Altman says ChatGPT's memory feature is an early step toward a model that learns about its user and becomes more useful over time.
“I think what people want, or at least what I want for myself, is a model that gets to know me and gets more useful to me over time.”
Sam Altman says OpenAI practises iterative deployment, publicising each GPT version instead of building GPT-5 in secret.
“rather than go build in secret until we got all the way to GPT five, we decided to talk about GPT one, two, three and four.”
Sam Altman used a rough estimate of 3 billion smartphone chips a year to argue that compute behaves unlike the phone chip market.
“Let's say they upgrade every two years, so the market per year is 3 billion system on chip for smartphones.”
Sam Altman says Helion is doing the best work on nuclear fusion, which he sees as part of the energy answer for AI compute.
“I think Helion's doing the best work, but I'm happy there's like a race for fusion right now.”
Sam Altman says the safest AGI scenario is a short timeline to AGI followed by a slow takeoff.
“I think short timelines, slow takeoff is the safest quadrant and the one I'd most like us to be in.”
Sam Altman says ChatGPT's subscription model means users pay and are not the product, which he prefers over ad-supported models.
“we have a very simple business model and I like it. And I know that I'm not the product, like, I know I'm paying”
Sam Altman proposes that AI companies publish the desired behavior of their models, so users can tell whether a failure is a bug or intended policy.
“it'd be nice to write out what the desired behavior of a model is, make that public, take input on it”
Sam Altman expects very capable AI systems by the end of the 2020s, possibly sooner, while calling the term AGI poorly defined.
“I expect that by the end of this decade, and possibly somewhat sooner than that, we will have quite capable systems”
Sam Altman says he has never had or wanted super-voting control over OpenAI, even after the board crisis.
“I continue to not want super voting control over OpenAI.”
Sam Altman says OpenAI finished training GPT-4 in the summer before its release and immediately gave it to outside red teamers to probe for harms.
“we finished last summer. Um, we immediately started giving it to people to, uh, to red team.”
With GPT-4, OpenAI introduced the system message, a way for users to steer the model's behavior that is separate from RLHF.
“We launched this thing with GPT four called the system message, um, which is not RLHF, but is a way to let users have a good degree of steerability over what they want.”
Sam Altman says the interview took place six days after GPT-4 launched.
“Today's Monday, we launched the previous Tuesday, so it's been six days.”
Sam Altman recounts an unverified story that Steve Jobs put a handle on the first iMac so people would know they could throw it out of a window.
“I don't know if it's true, I hope it is. Is that the reason Steve Jobs put that handle on the back of the first iMac”
Sam Altman compares the focus on neural-network parameter counts to the earlier gigahertz race among processor makers, arguing that real-world performance is what matters.
“people got caught up in the parameter count race in the same way they they got caught up in the gigahertz race of processors”
Sam Altman argues that chess is more popular than ever decades after Kasparov lost to Deep Blue, using this to suggest people will keep valuing human activities after AI surpasses them.
“I believe that chess has never been more popular than it is right now, and people keep wanting to play and wanting to watch.”
Sam Altman recommends a blog post by Eliezer Yudkowsky as the best steel man of the case that AI alignment is extremely hard, though he disagrees with much of it.
“he wrote this one blog post outlining why he believed that alignment was such a hard problem”
Sam Altman says he expected ChatGPT to be roughly the tenth fastest growing product in history, not the fastest.
“I thought it would've been like the 10th fastest growing product in history, and not the number one fastest”
Sam Altman says he considers consciousness strange but does not claim GPT-4 is conscious.
“I still think it's something very strange”
Sam Altman says he considers it certain that capable open-source large language models with little or no safety controls will soon exist.
“There are soon going to be a lot of capable open source LLMs with very few to none, no safety controls on them.”
Sam Altman says OpenAI was announced at the end of 2015 with a stated goal of working on AGI, and many people at the time thought that was absurd.
“announced the org at the end of 2015 and said we were gonna work on AGI, like people thought we were batshit insane”
Sam Altman says OpenAI began as a nonprofit but found early on that it needed far more capital than it could raise that way.
“we learned early on that we were gonna need far more capital than we were able to raise as a nonprofit”
OpenAI has a capped-profit subsidiary through which investors and employees earn a fixed return, with everything beyond the cap flowing to the controlling nonprofit.
“There is a subsidiary capped profit, so that our investors and employees can earn a certain fixed return, and then beyond that, everything else flows to the nonprofit.”
Sam Altman says OpenAI's profit cap started at 100x and is much lower for new investors.
“It started, is that it's much, much lower for, like, new investors now”
Sam Altman names customer service as a job category that AI could shrink substantially in the near term, though he says he is not certain.
“I guess I would say customer service is a category that I could see. There are just way fewer jobs relatively soon.”
Sam Altman says he helped start Worldcoin as a technological approach to the economic disruption AI may cause, and that OpenAI sponsored a large universal basic income study.
“I helped start a project called Worldcoin, um, which is a technological solution to this.”
Sam Altman says his working model is that the costs of intelligence and of energy will fall dramatically over the next couple of decades, making society much richer.
“the cost of intelligence and the cost of energy are going over the next couple of decades to dramatically, dramatically fall from where they are today”
Sam Altman says he still personally approves every hire at OpenAI and spends a large share of his time on hiring.
“I for real, truly do. Um, I still approve every single hire at OpenAI”
Sam Altman says Microsoft was unusual among large companies in understanding why OpenAI needed its special control provisions.
“They are unique in terms of the companies of that scale that understood why we needed the control provisions we have.”
Sam Altman says Silicon Valley Bank failed because it bought very long-dated instruments funded by short-term, variable deposits.
“buying very long dated instruments, secured by very short term and variable deposits”
Sam Altman suggests that a full guarantee of bank deposits, or a limit far above $250,000, would stop depositors from having to doubt their bank's safety.
“a full guarantee of deposits, maybe a much, much higher than 250 k”
Sam Altman says the Silicon Valley Bank run was much faster than the 2008 collapse because of Twitter and mobile banking apps.
“speed with which the SVB bank run happened because of Twitter, because of mobile banking apps”