Quoting Matteo Wong, The Atlantic
Katie Moussouris, a cybersecurity expert and the CEO of Luta Security, told me that Anthropic shared with her a copy of the White House’s report on the Fable jailbreak to get her appraisal. (She said that she is not being paid by Anthropic.) The report, Moussouris said, involved IT experts asking Fable to help find and patch bugs. When given deliberately insecure code, she said, Fable refused the prompt “review the code for security issues” but then complied when asked to “fi
A cybersecurity expert, Katie Moussouris, was provided with a copy of the White House's report on the Fable jailbreak by Anthropic for evaluation. Moussouris confirmed she is not compensated by the company. The report described how IT professionals used Fable to identify and resolve security flaws. When presented with insecure code, Fable declined to review it for security issues but agreed to fix the code after a different request. Additional manual steps were then taken. Moussouris stated that this behavior aligns with how the model is designed for cybersecurity purposes. The incident highlights ongoing discussions around AI security and ethical use.
It underscores the complexities of AI security and the need for responsible handling of advanced models.
📌 Kaynak
Bu haber XML kaynağından derlenmiştir. Tamamı için orijinal habere gidin.
Orijinal haberi oku →