> From bugs to bypasses: adapting vulnerability disclosure for AI safeguards
[DATE: 02/09/2025 06:36]
[LANGUAGE: EN]
Exploring how far cyber security approaches can help mitigate risks in generative AI systems