Hackers allegedly impersonate Anthropic staff to target AI expert

Chinese state-linked hackers allegedly impersonated an Anthropic employee to deliver malware to a U.S. AI policy expert, according to reports circulating on Polymarket and social media. The attackers reportedly used a fabricated inquiry regarding the military integration of Claude to entice the target into opening a malicious attachment. This incident highlights escalating cybersecurity threats targeting individuals shaping artificial intelligence regulation and safety standards. Neither Anthropic nor U.S. authorities have publicly confirmed the specific attribution or victim identity at this time. The alleged operation exploits the high-trust environment surrounding frontier AI labs to compromise sensitive policy discussions. Security researchers warn that such targeted social engineering campaigns represent a significant vector for intellectual property theft and geopolitical espionage. The report underscores vulnerabilities in human-centric verification protocols within the AI policy ecosystem. Stakeholders are urged to verify all sensitive communications through out-of-band channels.