Anthropic disclosed five instances in which researchers used Claude for work related to biological weapons, viruses, and toxins. The cases involved:
- Chikungunya virus—gain-of-function research to enhance transmissibility and immune evasion
- Highly pathogenic avian influenza—experiments to adapt it to mammals
- Orthopoxviruses (the family that includes smallpox and mpox)
- Venom peptides and toxins—computational redesign of dangerous proteins
Anthropic stressed that it’s not claiming the scientists intended harm. “The same information that can be used to develop a biological weapon could also be used to develop, for example, a vaccine or a cure for a disease”.
Surveillance and Espionage
The report also found cases involving:
- Iranian actors who developed a system capable of identifying people through their social media accounts
- A contractor working for Malian national security authorities who allegedly used Claude to develop software for an intelligence-gathering operation
- A Russia-linked cyber espionage campaign that used Claude to automatically rewrite malware to evade detection
Anthropic said these cases point to a broader shift in how AI can be used in state-linked operations. “AI is now being used in place of an engineering workforce”.
Laat een reactie achter