Given the current environment, many boards may be asking themselves whether they need to add directors with AI expertise in order to fulfill their oversight responsibilities with respect to their companies’ development and usage of AI tools. This Debevoise memo says that the answer to that question is more complicated than it seems at first glance:

While appointing a director with AI expertise may be appealing, it can present practical and governance challenges. First, the pool of individuals with both deep AI expertise and the qualifications to serve effectively as a public company director is limited.

Second, the percentage of companies for which AI is so fundamental to their business that it requires an AI expert on the board is very small. The appointment of a director with AI expertise could raise questions about a lack of specific board expertise covering other areas of potential enterprise risk (e.g., such as cybersecurity, political or environmental risks).

Third, the presence of a designated expert may inadvertently undermine effective board dynamics. For example, other directors may defer excessively to an AI expert, reducing the level of constructive challenge and debate that is critical to effective oversight. This dynamic can undermine the collective decision-making that is at the heart of board function and weaken the board’s ability to independently assess management’s approach to AI. Over time, concentrating AI knowledge in a single director may also reduce other directors’ incentives to learn about AI, which is likely to become increasingly important in the future.

Finally, individuals with deep AI expertise often have extensive experience in the technology industry and may have conflicts of interest, such as investments in AI companies or commercial relationships with vendors, which would require careful management.

The memo goes on to explain that adding an AI expert as a director isn’t the only way for a board to “get smart” about AI-related issues and discusses the role that expert guidances and appropriate education and regular reporting from management and outside advisors can play in supporting the board’s oversight of AI.

– John Jenkins