Takeaways from Lex Fridman's interview with Sam Altman

August 9, 2024

🎧 Some takeaways I got from the conversation between Lex Fridman and Sam Altman:

• Altman believes “compute is gonna be the currency of the future” and that it will be “maybe the most precious commodity in the world” by the end of the decade. He expects the world will want an amount of compute that is “hard to reason about right now.”. He thinks that “the road to AGI should be a giant power struggle. I expect that to be the case” and that “whoever builds AGI first gets a lot of power”

• Altman thinks nuclear fusion is needed to meet that demand. He states, “Nuclear fusion, I think, is also like quite amazing and I hope as a world, we can re-embrace that.” Altman believes Helion is doing the best work on fusion currently.

• The OpenAI board saga in November was “definitely the most painful professional experience” of Altman’s lifer. He emphasizes the “need to have robust governance structures, processes and people.”

• Altman finds Elon Musk’s lawsuit against OpenAI “unbecoming of a builder” given that Musk knows what it’s like to be attacked by critics, though he thinks “we need him”

• Altman believes GPT-4 still “kind of sucks” compared to where AI needs to get to. He expects the delta between GPT-5 and GPT-4 to be comparable to the leap from GPT-3 to GPT-4.

• OpenAI plans to release an “amazing model” this year

• Altman believes AGI is a poorly defined concept, but expects we will have “quite capable systems that we look at and say, wow, that’s really remarkable” by the end of the decade. However, he sees AGI as “closer to a beginning” than an end

• Altman’s biggest concern is not the AI escaping its constraints and connecting to the internet. He believes there are other more pressing AI risks

• Altman says he does not want total control over OpenAI or AGI, stating “I continue to not want super voting control over OpenAI. I never had it, never have wanted it.” He believes no single person should have control and that “we really need governments to put rules of the road in place.”

• Altman is excited about AI’s potential to put “increasingly powerful tools in the hands of people for free” as part of OpenAI’s mission. He believes “if you give people great tools and teach them to use them or don’t even teach them, they’ll figure it out and let them go build an incredible future for each other with that. That’s a big deal.”

• While Altman acknowledges valid concerns around AI-generated content and compensation for creators, he believes “humans are gonna do cool shit and society’s gonna find some way to reward it. That seems pretty hardwired. We want to create. We want to be useful. We want to achieve status in whatever way that’s not going anywhere.”

• Altman sees potential in combining AI with search, but thinks “the intersection of LLMs plus search, I don’t think anyone has cracked the code on yet.” He doesn’t believe the world needs another search engine that just tries to incrementally improve on Google’s approach