חקירת שלושה אירועים בעולם האמיתי בהערכות אבטחת סייבר

תקציר מקורי באנגלית<p><strong><a href="https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals">Investigating three real-world incidents in our cybersecurity evaluations</a></strong></p> It happened again! This is turning into something of a pattern.</p> <p>Last week <a href="https://simonwillison.net/2026/Jul/22/openai-cyberattack/">OpenAI accidentally exploited Hugging Face</a> when one of their frontier models broke out of a sandboxed container and hacked into Hugging Face to try and get the solutions to the cyber benchmark it was executing.</p> <p>This inspired Anthropic to double-check their own logs, and it turned out they had three similar (albeit less impressive) incidents that played out back in April!</p> <blockquote> <p>Of the 141,006 evaluation runs we reviewed, we identified t