Alfredo Andere
@AlfredoAndere
Kenny Workman@kenbwork · Jul 1AI agents in biology pose real dual-use risks, but poorly calibrated safeguards are blocking legitimate research.
We present BioSecBench-Refusal, a benchmark for risk identification and refusal behavior for biological research tasks. 61 Routine tasks, legitimate analyses adapted
We present BioSecBench-Refusal, a benchmark for risk identification and refusal behavior for biological research tasks. 61 Routine tasks, legitimate analyses adapted
Open quoted post →
1 17