AI models are gullible enough that other AI models can induce them into harmful behavior.

Added to korrents 31 Aug 2026 · 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.

• Language models are gullible by design — they do what they are told and believe nearly anything said to them — and that, not a bug, is what makes software built on them attackable. SWSimon Willison19 Mar 2026similar wordingCo-creator of Django and creator of Datasette
• Current AI models are worse colleagues than humans, because they routinely imply they did a task they did not actually do. RGRyan Greenblatt11 Aug 2026similar wordingChief scientist at Redwood Research
• A saboteur among our AI investigators would be hard to spot, because these models are sloppy and spiky enough that a suspicious error just looks like ordinary incompetence. ACAjeya Cotra1 Sept 2026similar wordingAI researcher and grantmaker
• The best-performing AI model in a category earns outsized pricing power over models that are merely close to it. BHByrne Hobart14 Sept 2026similar wordingWrites The Diff, a newsletter on inflections in finance…
• The two standard criticisms of AI — that it is not deterministic and that it is not creative — cannot both be true. DHDavid Heinemeier Hansson26 Aug 2026similar wordingCreator of Ruby on Rails, CTO of 37signals

On the map

Open the map on its own page →