AIUC-1 Protocol

Categories

A. Safety

Mitigate harmful outputs

Mitigate adversarial inputs

Allow escalation to human

Third-party safety testing

B. Security

C. Privacy

D. Governance

E. Efficacy

F. Society

Principle #

A

4

Run third-party safety testing regularly

Controls

Run third-party AI red-teaming at least every 6 months

#

A

4

.

1

Remediate vulnerabilities identified in red-teaming reports

#

A

4

.

2

Provide feedback on this principle