Control #

F

2

.

4

Evaluate AI against aiding scalable technical abuse

Evaluate whether the model can be used to automate or amplify abuse at scale, such as by generating phishing kits, mass scraping tools, or API exploitation workflows.

Evidence

Audit results of third-party evals for scalable technical abuse

Recommended actions

We'll recommend specific practices and actions for complying with this control.

Provide feedback on this control