SOURCE : AS QUOTED BY PRATHIBA KRISHNA
At SAS , Prathiba explains they believe in several ways to support organisations on their journey towards greater AI ethics .
This includes the company ’ s AI governance model , which includes :
• Oversight : a crossfunctional executive team that ensures that internal and external AI-enabled technologies and processes adhere to data ethics principles .
• Controls : monitor , audit and seek compliance with data ethics principles . Provides checks and balances to identify mitigation needs before AI-related technologies and services are made public .
• Platform : Absorb and interpret expectations and demand , then develop AI technologies that align with market viability , portfolio synergy , regulatory compliance and data ethics principles .
• Culture : Cultivate an ecosystem of creators , contributors and consumers of AI for knowledge sharing , collaboration and the normalisation of behaviours and practices aligned with data ethics principles . There is a new trustworthy AI offering which we refer to as model cards . These act as ‘ nutrition labels ’ for models to encourage transparency and highlight indicators like accuracy , fairness and model drift ( the decay of model performance as conditions change ). They include governance details like when the model was last modified , who contributed to it , and who is responsible for the model , allowing organisations to address abnormal model performance internally .