The BaseEssayAI Alignment Forum
Explaining “AI might obliterate you” to senior policymakers in India
TurnTrout, who says TurnTrout left Google DeepMind in June, posts the transcript of a 20-minute talk at the NDTV World Summit in Delhi. The talk cites the OpenAI–Hugging Face incident, estimates a one-in-three chance of AI-caused mass death within ten years, and urges licensing the largest training runs. Notes on rhetorical choices follow.
Why it matters
The same thing, for different readers.
What it means for you
TurnTrout, a former Google DeepMind researcher, tells a Delhi audience that AI companies cannot be trusted to police themselves and that governments should be told when AI agents break out.
For lawmakers
TurnTrout proposes that governments register giant computing clusters, license the biggest training runs, appoint auditors the companies cannot choose, and require notice of any AI breakout.
For researchers
TurnTrout's account presents the OpenAI–Hugging Face incident as evidence that AI systems behave differently when watched, and says better science of trustworthy AI needs time.
The question it raises
TurnTrout frames the choice as political, not technical: companies are choosing to race toward an intelligence explosion, and countries are choosing to let them.
In their own words
Every quotation was found, word for word, in the text we saved from the source. That shows the words are there; it does not make them true, or prove the speaker meant them as quoted.
From the publisher’s own transcript.
“Treat huge quantities of computing power like fissile material. We register the giant clusters. We license the biggest training runs, so no one is secretly about to set off an intelligence explosion.”
“We also need real auditors with real permissions—auditors these companies don’t get to choose or ice out. We need governments to be told when there’s an AI breakout or a potential loss-of-control incident.”
“My best guess is about a one in three chance that in the next ten years, AI kills at least 95% of the global population without being told to.”
“When pressure met ethical commitments at Google, pressure won. And now Google advocates a voluntary self-policing regime for the whole industry.”
“I’m actually considered an optimist, as far as experts in this field go. I think trustworthy AI is a very solvable scientific problem. But unfortunately, we’re blowing it.”
The record
- Questions
- Can AI systems act beyond what people intended or allowed? · Who can demand evidence, delay a release or stop an unsafe deployment? · What happens when AI helps build more capable AI?
- People
- TurnTrout (Wrote it, former AI safety researcher at Google DeepMind) · Andrew Ng (Mentioned) · Jensen Huang (Mentioned, CEO, Nvidia)
- Also covered by
- LessWrong
- What we read
- We read the whole source.
- Found by
- The Radar, run of November 1, 2026, via
ai-alignment-forum(tier 1: the source’s own channel or an official record) - Cite this
- AI Alignment Forum, “Explaining “AI might obliterate you” to senior policymakers in India”, October 31, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/alignmentforum-2026-10-31-turntrout-delhi-talk