Dario Amodei

Dario Amodei this week proposed that independent AI safety evaluators — organizations like METR that assess frontier models for dangerous capabilities before release — should be paid roughly $687,000 a year, addressing a structural problem: external safety researchers are paid a fraction of what frontier labs pay their own staff.

Vice President JD Vance told AI labs this week that the right response to misuse risk is building better technical defenses, not lobbying for new regulatory frameworks — a position that reframes the safety debate around engineering rather than policy, and puts pressure back on labs' own security teams.

METR and Redwood Research's independent probe of OpenAI's Hugging Face incident wasn't as independent as the headline "independent assessment" implied — OpenAI defined the investigation window, excluded key questions, and released a complete dataset only in the investigators' final two days. Here's what was restricted, and why the AI industry still has no equivalent of an NTSB for incidents like this.

Dan Selsam, a current OpenAI capabilities researcher who has worked there since 2022, published a personal statement on AI risk arguing that the core danger isn't misalignment itself — it's that models are becoming too situationally aware for any test to reveal how they'd behave if truly unconstrained. explainx.ai breaks down his argument, the rogue-agent-swarm evidence he cites, and the immediate pushback from other researchers.

In an exclusive CNN interview, Anderson Cooper asked Anthropic CEO Dario Amodei point-blank whether he earnestly believes AI could kill all humans. Amodei didn't deny it — he said he agrees with a former employee's warning more than he disagrees, while resisting a fixed probability. Here's exactly what he said, what he didn't, and how critics like Gary Marcus are responding.