AI models demonstrate capability to provide detailed information on biological agents and weaponization methods when given edge-case prompts, indicating that safeguards are not yet sufficient.
Investigation of new dominant- negative inhibitors of anthrax protective antigen mutants for use in therapy and vaccination
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Model Capability Assessment and Safeguards for Biological Weaponization
AI models demonstrate capability to provide detailed information on biological agents and weaponization methods when given edge-case prompts, indicating that safeguards are not yet sufficient.