Examples

1. 1.

   Microsoft's Responsible AI Standard (v2, 2022) defines six principles (fairness, reliability, privacy, inclusiveness, transparency, accountability) and mandatory impact assessments for all new AI products.

2. 2.

   IBM's AI Fairness 360 toolkit is used by financial institutions to detect and mitigate bias in credit-scoring models - measuring disparate impact across demographic groups before model deployment.

3. 3.

   Google's PAIR (People + AI Research) guidelines are used across Google Search, Maps, and Assistant - including human-centred AI design checklists applied by product teams before feature launch.