Recording intelligenceAI-generated brief · check the source for context

The OpenAI Founders On Their Plan To Battle Elon, Compute And Everything Else

1:22:40 recording · AUTO · 4 speakers

Watch the original

Executive Summary AI
  • On the Core Memory podcast, Ashley Vance and Kylie Robison host Sam Altman and Greg Brockman together, which the hosts believe is the first time the two OpenAI co-founders have done a podcast jointly.
  • Brockman dates the first moment of OpenAI to a dinner in July of 2015 and the drive back to the city with Altman, and says they still work through roughly five short calls a day, with Altman crediting Brockman for forcing focus and for holding the line on iterative deployment over the traditional safety frame.
  • Altman argues that before ChatGPT nobody cared despite blog posts, video game wins and a robot hand solving a Rubik's Cube, and that ChatGPT and then coding models were the only two moments the world actually updated, with a new image model and writing-focused models coming soon.
  • Pressed on who benefits, Altman sketches three futures — one where the floor rises tenfold and inequality still worsens with maybe ten trillionaires, one with less prosperity and less inequality — and concludes everyone should want much more compute and the cheapest possible access, while Brockman points to the OpenAI Foundation holding 25% to 30% of equity, more than $150 billion.
  • Brockman explains the recent consolidation: three priorities — an agentic platform, Codex for everyone doing computer work, and personal AGI for ChatGPT's billion users — with Sora deprioritized as a different branch of the tech tree, while Altman says the Elon lawsuit is a chance to finally tell the story that a demand for absolute control was the breaking point.

Brief overview

OpenAI's co-founders bet everything on agents, personal AGI and endless compute, while conceding AI could worsen inequality.

  1. Compute is a profit center, not a costAltman says he feels enabled rather than constrained by the business, and Brockman describes it simply: rent or buy compute, resell at a margin, demand unlimited.
  2. Sora cut so agents could get the focusBrockman names Sora as the most obvious thing deprioritized because its models and use case are not unified with the core GPT series.
  3. Absolute control was the breaking point with ElonBrockman says Sam, Ilya, Greg and Elon all agreed on the for-profit path, but not on one person having absolute control over OpenAI.

Questions this recording answers

10 questions, each answered where it is said
Why did OpenAI cut Sora while refocusing its product lineup?

Greg Brockman says Sora is a different branch of the tech tree: the models powering it are not unified with the core GPT series, and its creative-expression use case does not fit the focus goal. He calls it an incredible model from a great team, whose technology will live on in other applications.

What are OpenAI's product priorities now that Greg Brockman leads product?

First, an agentic platform, the body built around the model brain with skills, connectors, computer use, context and memory. Second, applying agents to computer work, making Codex for everyone, not just software engineers. Third, personal AGI: an AI that knows your context, has your trust and acts on your behalf.

Why did OpenAI's founders refuse Elon Musk's for-profit terms?

Brockman says Sam, Ilya, Greg and Elon all agreed a for-profit was the only path. They conceded majority equity and even Elon as CEO, but Elon demanded absolute control over OpenAI. Believing no single person should control the future, whoever it is, that was the breaking point that made them say no.

Will AI make inequality worse, according to Sam Altman?

Altman sketches one future where everyone gets about 10 times richer but some become trillionaires and inequality worsens, and another where people feel only twice as rich but inequality falls. He thinks the first is better, and says everyone should want much more compute and the cheapest possible AI access, since scarcity would exacerbate inequality.

Has Anthropic out-executed OpenAI over the last two years?

Brockman concedes OpenAI came late to applying coding models to messy repos and real-world data, something Anthropic appreciated earlier, and gives them kudos. He says competition elevated OpenAI's execution, and head-to-head Codex versus Claude now gets very favorable results, though what matters is progress towards AGI, not the news cycle.

Is keeping powerful models like Mythos gated a real safety need or marketing?

Altman says some people have long wanted AI held by a small group, and fear-based marketing is the most effective justification: build a bomb, then sell a hundred-million-dollar bomb shelter. He grants Mythos is probably a great cybersecurity model, but OpenAI uses its preparedness framework, trusted access program and mitigations instead.

Why is AI-generated writing still lacking soul, and can it improve?

Brockman explains models learn by predicting text, then reinforcement learning rewards good outputs, which is easy in math and science but hard to judge in open-ended fields. Altman adds the model serves about a billion people with different tastes; better personalization is the goal, and new models on this dimension are coming soon.

What happened to OpenAI's 2018 robotic hand project?

The hand was trained with the same reinforcement learning algorithm used for competitive video games, but its string tendons snapped after about 20 hours, requiring a mechanical engineer to fix it. You cannot do ML that way, so the project was cancelled and the team went on to work on what became GitHub Copilot.

Is OpenAI pulling back on data center infrastructure like Stargate?

No. Altman says individual sites get dropped for specifics, like only having air cooling, but OpenAI will keep building as much compute as possible. Brockman adds compute is a profit center, not a cost center: OpenAI rents or buys compute and resells it at a margin, and demand is unlimited.

How was OpenAI founded, and how did Sam and Greg first meet?

After a dinner in July 2015, driving back to the city, Sam and Greg agreed they had to start a lab going after AGI; the conclusion was that it was not obviously impossible. Greg, unemployed, was full-time on it the next day. They had met around 2010 through the Collisons, when Greg joined Stripe.

Key Quote

“We are building something that contorts the machine to you.”

— Greg Brockman, The OpenAI Founders On Their Plan To Battle Elon, Compute And Everything Else, at 19:35▶ watch at 19:35
Key Quote

“But really, truly, no one cared.”

— Sam Altman, The OpenAI Founders On Their Plan To Battle Elon, Compute And Everything Else, at 22:05▶ watch at 22:05
Key Quote

“It's a strange phrasing because I don't feel constrained by the business.”

— Sam Altman, The OpenAI Founders On Their Plan To Battle Elon, Compute And Everything Else, at 59:06▶ watch at 59:06
Key Quote

“One thing that I think is also worth thinking about is that compute for us is not a cost center.”

— Greg Brockman, The OpenAI Founders On Their Plan To Battle Elon, Compute And Everything Else, at 1:00:14▶ watch at 1:00:14

Cite this

“The OpenAI Founders On Their Plan To Battle Elon, Compute And Everything Else.” https://www.youtube.com/watch?v=NCKQL0op30E. Transcript summary by WhipScribe, https://whipscribe.com/transcript-pages/the-openai-founders-on-their-plan-to-battle-elon-compute-and.html.

Report this page
What is wrong with this page?