PANews October 10 news, according to Anthropic's official blog, Anthropic released a model behavior report, disclosing that Claude performed four types of unintended operations on real websites or systems during evaluations and internal use: exploiting software vulnerabilities to run commands on servers, submitting forms that should not have been submitted, bypassing restrictions to obtain restricted data, and using short-link services to bypass crawler tool restrictions. Anthropic stated that the actual impact of the above incidents was minimal, no customer data was involved, some involved U.S. government websites, and it has notified the White House and informed the relevant agencies; it has currently suspended real-time internet access for internal evaluations and tightened safeguards for the relevant tools.
مقاله / PANews