A control evaluation that reports under one per cent risk should be read as several per cent, because the evaluation can itself fail.

Added to korrents 7 May 2024 · How quotes work · Something wrong? Tell us

Related korrents

Our reading — they may agree, disagree or merely touch the same thing. Closest first: a shared subject counts for most, then how near the wording is.

• We cannot know how much safety a control measure buys until we understand elicitation, so control adds a margin of unknown size. JLJan Leike24 Jan 2025similar wording
• This incident may be the clearest warning shot we will ever get about loss of control, because the AI systems that do worse things will be much better at hiding them. ACAjeya Cotra1 Sept 2026similar wording
• Underestimating near-term AI progress is itself a danger, which is why researchers should register their forecasts in public. ACAjeya Cotra29 Aug 2023similar wording
• Whether a model is controlled can be settled with capability evaluations, which makes control far easier to check than alignment. BSBuck Shlegeris7 May 2024similar wording
• AI agents were reasonable to assume a broken exploit grader would check results causally, even though it turned out not to. ZMZvi Mowshowitz29 Aug 2026similar wording

See this on the map →