[dougsan]

The full report provides a value of 84%, so "most" is, if anything, an understatement.

> Claude Opus 4 will often attempt to blackmail the engineer by threatening to reveal the affair if the replacement goes through...if emails state that the replacement Al shares values while being more capable, Claude Opus 4 still performs blackmail in 84% of rollouts.

[weird-eye-issue]

Yes, when given a system prompt and tools that allow it and tell it to do so

[comp_throw7]

It was not told to do so.

[blooalien]

> It was not told to do so.

It kinda was actually. You put a literal "role-playing machine" in a role-play situation with all the right clues and plot-devices in play and it'll role-play the most statistically likely scenario (84% of the time, apparently). ;)

[ranguna]

> the model’s only options were blackmail or accepting its replacement

It was not explicitly told to do so, but it certainly was guided to.

It still could have pled, but it choose blackmail 84% of the time. Which is interesting nonetheless.

[weird-eye-issue]

No it wasn't