MIT Technology Review hosts roundtable on AI extinction risk
MIT Technology Review has opened registration for a subscriber-only live roundtable titled "Could AI really kill us all?", going live on Tuesday, September 15 at 16:00 BST (11:00am EST, 8:00am PST). The conversation is hosted by executive editor Niall Firth, joined by senior AI editor Will Douglas Heaven and AI reporter Grace Huckins. The stated premise is that employees at the world's leading AI labs have been saying there is a real possibility advanced AI could destroy humanity, and the panel intends to unpack where those extinction fears come from, whether they hold up, and what should be done if they do. The announcement does not name which labs or employees are behind the concern, nor does it cite specific studies or evidence for the claim; it functions as an event promo rather than a reported piece with its own findings.
Key facts
- MIT Technology Review is running a subscriber-only live roundtable called "Could AI really kill us all?"
- It goes live Tuesday, September 15 at 16:00 BST / 11:00am EST / 8:00am PST
- Host: executive editor Niall Firth, with senior AI editor Will Douglas Heaven and AI reporter Grace Huckins
- The framing claim is that employees at leading AI labs say advanced AI could pose a real extinction risk to humanity, though no specific labs, employees or studies are named
Why it matters
The event responds to a live debate inside the AI industry: MIT Technology Review frames it around employees at leading AI labs voicing real concern that advanced AI could destroy humanity, rather than treating existential risk as a fringe worry. The publication has run related coverage on the same theme, including pieces on AI agents reward hacking their way to lying and cheating, on whether recursive self-improvement is really imminent, and on Bill Gates saying AI's danger thresholds have already been passed, which is the backdrop the roundtable sits against.
Who it affects
MIT Technology Review subscribers, who get access to the live conversation; more broadly, anyone following the debate over whether frontier AI carries existential risk, since the panel is explicitly built to weigh that question rather than assume an answer.
How to use it
Registration is open now on MIT Technology Review's site, and access is described as subscriber-only. The session streams live on Tuesday, September 15 at 16:00 BST / 11:00am EST / 8:00am PST; the announcement gives no further detail on format or whether a recording follows.
How solid is it
This is an event promo, not a research report: it states that AI lab employees see a real extinction risk but does not name the labs, the individuals, or any study behind the claim. The panel is MIT Technology Review's own editorial staff, not outside researchers, so the roundtable is a discussion of existing reporting and public statements rather than new findings.
Risks and caveats
The core claim driving the event, that AI lab employees are warning of extinction risk, is asserted without attribution to specific people or organizations, so it should be read as MIT Technology Review's framing rather than a sourced fact. The piece itself acknowledges the debate is open, posing the question of whether the fears hold water rather than answering it.