Open priamai opened 1 year ago
Please, refer to this work for ML threat mitigations: https://arxiv.org/pdf/2207.00091.pdf. We are also implementing an AI SOAR feature to help prevent vulnerabilities in AI tools and mitigate against AI threats (see CertEye EOAR feature) : https://oecd.ai/fr/catalogue/tools/certeye and https://github.com/ngatilio/CertEye.
Lionel Tidjon, Ph.D Founder, certkor.ai
Hi there, one thing I didn't see in the taxonomy SEP, an optional reference field that would allow mitigations to be considered. This would be similar to the Defend approach.