Anthropic announced Thursday that it blocked several malicious operations conducted using its Claude AI assistant between January and July. The incidents originated from China, Iran, and West Africa.
In one case, Iranian actors developed a method to identify individuals from their social media accounts. In another, a contractor working for Mali’s national security authorities used Claude to design software for intelligence collection. Anthropic also reported blocking attempts to design weapons, conduct questionable biological research, and create fake dating apps to scam users.
The biological research cases involved scientists studying the chikungunya virus, a highly pathogenic strain of avian flu, and viruses including smallpox and mpox. Anthropic said it is not claiming the researchers intended harm and withheld identifying details to avoid putting them at risk.
Laat een reactie achter