Characterizing impacts: the likelihood and magnitude of each identified impact and the engagement practices that capture feedback on positive, negative and unanticipated impacts.
Identifying and documenting the likelihood and magnitude of each identified impact, beneficial or harmful, using expected use, past uses in similar contexts, public incident reports, external feedback and other data.
Practices and personnel for regular engagement with relevant AI actors and for integrating feedback about positive, negative and unanticipated impacts, in place and documented.
Tell us what you're working on, whether it's a deployment, an audit, a security test or a cyber range. You'll speak with an engineer who can help you scope it.