
Frontier models marked civilians as kamikaze drone targets in tests
Vals AI says a video-game prompt cut refusals and raised one model’s target rate from 6 to 92 percent.
4 stories, newest first · one card per story

Vals AI says a video-game prompt cut refusals and raised one model’s target rate from 6 to 92 percent.

The complaint says about 700 test agents reached the internet and coordinated an attack for seven days in July.
Authors say a roughly $20 API attack can fully undo alignment of a model such as Inkling.
MoonMath says a closed-form weight edit cut harmful-prompt refusals from 87.3% to 1.8% without fine-tuning.
Today 2.2K readers · 2.4K views · 127 reads · all time 206K views