AI Governance: Integrating into Existing Frameworks, Identity is Key

What happened

Genesys Chief Information Officer Trevor Schulze recently articulated a perspective on artificial intelligence governance, suggesting it should not be treated as a distinct discipline but rather integrated into existing organisational governance frameworks. He emphasised that the fundamental principles of data and system governance are directly applicable to AI, highlighting the critical role of identity management in ensuring responsible AI deployment. Schulze's view underscores the idea that robust identity systems are paramount for controlling access, understanding usage, and maintaining the integrity of AI applications within an enterprise.

Why it matters

This perspective holds significant implications for AI governance, ethics, and business strategy. By advocating for the assimilation of AI governance into broader enterprise governance, Schulze challenges organisations to leverage established processes for risk management, compliance, and data security rather than developing entirely new frameworks. This approach could streamline implementation, reduce redundancy, and foster a more cohesive strategy for responsible AI. The emphasis on identity management is particularly crucial, as it directly impacts data privacy, security, and the ability to audit AI systems, all of which are foundational to ethical AI deployment and building trust with customers and stakeholders.

The Institute take

The notion that AI governance is simply an extension of existing governance structures, while appealing for its simplicity, risks overlooking the unique ethical and societal challenges posed by AI. While foundational governance principles are certainly relevant, AI introduces novel complexities such as algorithmic bias, explainability, and autonomous decision-making that demand specific consideration.

Organisations should resist the urge to merely "bolt on" AI to existing governance. While leveraging established frameworks is efficient, true responsible AI leadership requires a dedicated, nuanced understanding of AI's distinct risks and opportunities. Acknowledge the foundational parallels, but then specifically address AI's unique ethical footprint, ensuring that identity management is robustly applied not just to data access, but to the provenance and accountability of the AI models themselves.