Responsible AI Checklist for Managers

Name the owner

• Assign a named owner for each material AI use.
• Record purpose, affected users and the decision it influences.
• Set a date for review rather than approving a tool forever.

Bound the system

• Specify approved data sources and prohibited inputs.
• Limit permissions to what the task requires.
• Define which outputs are advice and which can trigger action.

Check people and impact

• Identify who may be disadvantaged by errors or bias.
• Give users a way to challenge consequential outcomes.
• Make human review meaningful rather than ceremonial.

Monitor and stop

• Track failures and complaints as well as success metrics.
• Reassess after supplier or model changes.
• Keep a practical kill switch or fallback process for material failures.

Evidence behind this checklist

• European Commission, AI Office · AI Literacy - Questions & Answers · 2025-11-19 · Primary source
• Information Commissioner’s Office · Guidance on AI and data protection · 2025-06-19 · Primary source
• UK Government · Code of Practice for the Cyber Security of AI · 2025-01-31 · Primary source

How evidence is selected and checked