Safety3d ago
Safety
LLMs Struggle With Unclassified Biological Threats
Mirror RNA polymerase creation in 2022 highlights a critical gap in AI safety filters. Current models rely on official CBRN classifications, often failing to refuse harmful requests regarding emerging risks like mirror life. This lag between scientific discovery and policy updates leaves a window for misuse. Researchers must develop safety systems that anticipate threats before regulators codify them.