Assess whether an agent may take actions without a human gate (05fb4e)

Situation

After a journalist asking if the hiring tool is biased, EU AI Act high-risk classification worksheet is what chief AI officer can touch in a retailer using generative AI in customer service. AI Governance will live with Policy or governance breach versus Model defect on this Vendors and Agentic Systems file.

Decision

Chief AI officer in a retailer using generative AI in customer service must choose Policy or governance breach / Model defect / Dual failure / Hold for the missing fact using EU AI Act high-risk classification worksheet after a journalist asking if the hiring tool is biased.

Hypotheses to test

• Chief AI officer can defend Policy or governance breach from EU AI Act high-risk classification worksheet after a journalist asking if the hiring tool is biased in a AI Governance challenge.
• Chief AI officer cannot defend Policy or governance breach from EU AI Act high-risk classification worksheet; Model defect is what the extract actually supports after a journalist asking if the hiring tool is biased.
• A journalist asking if the hiring tool is biased never reached the population in EU AI Act high-risk classification worksheet — reopen intake, do not close an agent may take.
• Two facts in EU AI Act high-risk classification worksheet after a journalist asking if the hiring tool is biased conflict for chief AI officer; hold this Vendors and Agentic Systems file.

Analysis required

• Split policy-or-governance failure from a model defect using prompts, outputs, and human edits in EU AI Act high-risk classification worksheet.
• Reproduce the incident row in EU AI Act high-risk classification worksheet and say whether it ever touched production data.
• Split policy-or-governance failure from a model defect using prompts, outputs, and human edits in EU AI Act high-risk classification worksheet.
• For this AI Governance Vendors and Agentic Systems file, read EU AI Act high-risk classification worksheet against a journalist asking if the hiring tool is biased and write the one fact that would move an agent may take for chief AI officer.

Recommendation

Choose Policy or governance breach / Model defect / Dual failure / Hold for the missing fact on this AI Governance / Vendors and Agentic Systems packet (EU AI Act high-risk classification worksheet after a journalist asking if the hiring tool is biased). The follow-on Vendors and Agentic Systems action is what chief AI officer does next: implement the option, assign an owner, and log the missing fact.