AI developer Anthropic has published a report documenting instances where users attempted to leverage its Claude models for harmful purposes. The company identified efforts to utilize its technology for cyberattacks, surveillance, influence operations, and the development of biological weapons. According to the firm, these attempts were successfully detected and mitigated by their security systems. Anthropic stated that the findings from these incidents are being used to strengthen safety protocols and improve future defensive measures. The release of this information follows growing international scrutiny regarding the potential risks associated with the increasing capabilities of advanced artificial intelligence systems and their impact on public safety.
Source : The National
Photo : The National
