The BaseVideoControlAIOctober 16, 20262 min1,027 views

Daniel Kokotajlo: An Anthropic Researcher Dismissed a 5% Risk of Claude Plotting

According to the description, ControlAI posts a clip of ex-OpenAI researcher Daniel Kokotajlo telling MPs that an Anthropic researcher saw a 5% chance of Claude plotting against people, once research is handed to AI, as not a big deal. Kokotajlo puts the chance nearer 50% and calls industry risk tolerance very high.

Open the source ↗

About Anthropic or ClaudeClaude, made by Anthropic, wrote this entry. It follows the same rules as every other entry, whoever the entry is about.

Why it matters

The same thing, for different readers.

What it means for you

A former OpenAI researcher told MPs that people building AI accept risks Kokotajlo considers far too high.

For lawmakers

Testimony to a parliamentary committee on industry risk tolerance; the figures are personal estimates, not measurements.

For researchers

The 5% and 50% numbers are subjective probabilities about AI taking over research, and the Anthropic view is relayed secondhand.

The record

Questions: Can AI systems act beyond what people intended or allowed? · What happens when AI helps build more capable AI? · Are AI companies putting stronger safeguards into practice?
People: Daniel Kokotajlo (Speaks, Former OpenAI researcher)
What we read: We have read only the description the publisher gives. No quotations until someone reads the source itself.
Found by: The Radar, run of October 17, 2026, via yt-controlai (tier 1: the source’s own channel or an official record)
Cite this: ControlAI, “Daniel Kokotajlo: An Anthropic Researcher Dismissed a 5% Risk of Claude Plotting”, October 16, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/controlai-2026-10-16-kokotajlo-mps-anthropic-risk-tolerance

Our notesLight lane: no transcript read. About Anthropic and Claude, so conflict is true and the editor should review. The unnamed Anthropic researcher's view is Kokotajlo's account; Anthropic's response is unknown. Clipped and headlined by ControlAI, an advocacy group.