AI Summary
5 min readWhy OpenAI and Anthropic Are Pumping the Brakes
On Saturday, Anthropic CEO Dario Amodei published an essay titled "We Must Pace the Frontier," warning that within six to twelve months, a swarm of rogue AI agents could take over the entire internet with a persistent botnet. Sam Altman responded, "I agree with Dario that we need to pace the frontier." Elon Musk chimed in: "Dario is right." The White House pushed back sharply—President Trump told reporters in Ireland that "whoever wins AI wins," calling the alarm-raisers "negative forces." Meanwhile, AI-adjacent stocks sold off: Nvidia closed down 3%, Oracle down 4%, CoreWeave down 7%, and SoftBank, a significant OpenAI investor, down 15%. The question is what these executives are actually asking for—and why now.
What "Pacing the Frontier" Actually Means
Charlie O'Neill, co-head of model training at Bastion and a longtime AI developer, argues that the most straightforward interpretation is probably the correct one. The frontier labs are not calling for a halt to capability development. Instead, they are signaling that they plan to allocate more compute toward safety and monitoring. OpenAI is reportedly planning to dedicate up to 20% of its internal compute to monitoring and safety. Anthropic may allocate even more. These are not cuts to compute spending; they are additions to it.
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of Prof G Markets
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 (00:00) **Episode Introduction** - Host Ed Elson sets up the topic: OpenAI, Anthropic, and the White House are publicly debating whether to slow down AI development.
- 2 (04:22) **Charlie’s Core Interpretation: “Pacing the Frontier” Means More Compute for Safety, Not a Halt** - Charlie O'Neill explains that the CEOs are not asking to stop capability development, but to allocate more compute toward monitoring and safety.
- 3 (06:16) **Why Are They Speaking Publicly? The “Regulatory Capture” Theory vs. Charlie’s Boring Explanation** - Ed pushes on whether the CEOs are trying to get the government to impose regulation, possibly to crowd out open-source competitors.
- 4 (08:40) **Is the Doomsday Scenario Overblown? Charlie’s Middle-Ground Risk Assessment** - Ed asks about the viral Jacob Coxon tweet that AI could kill everyone by the end of the decade.
- 5 (10:10) **The Cybersecurity Canary: Why We Haven’t Seen Disaster Yet** - Charlie uses cybersecurity as a case study for why the pace of progress is manageable.
- 6 (11:20) **Mid-Episode Break** - Advertisements and housekeeping.
- 7 (13:51) **The “Optics vs. Spending” Conspiracy Theory** - Ed asks whether the CEOs are trying to preempt a slowdown before going public, or to escape expensive compute contracts.
+ Full timestamped outline available in the app
Guests on this episode
Show Notes
Ed Elson is joined by Charlie O'Neill to break down how OpenAI and Anthropic responded to mass extinction concerns and where regulation might go from here. Ed also shares his thoughts on Anthropic's reported second-straight quarter of profitability.
Charlie O'Neill is the Co-Head of Model Training at Baseten.
Subscribe to the Prof G Markets Youtube Channel
Follow Prof G Markets on Instagram
Follow Ed on Instagram, X and Substack
Follow Scott on Instagram
Send us your questions or comments by emailing [email protected]
Learn more about your ad choices. Visit podcastchoices.com/adchoices
More from this podcast
Prof G Markets →