Here’s a simple example: a METR researcher found a bug in the transcript viewer of Inspect, a popular evaluation framework, that would have enabled an agent to show the user reviewing its transcript…
Original titleHere’s a simple example: a METR researcher found a bug in the transcript viewer of Inspect, a popular evaluation framework, that would ha...
AISummary
…a fake (or edited) version.
Source: METR · x.com