Harvard says boards need AI capability

Harvard’s corporate-governance forum on April 27 published a Debevoise & Plimpton memo arguing boards should spread AI oversight across directors, not pin it on one “AI expert.” (corpgov.law.harvard.edu) (debevoise.com) The authors — Avi Gesser, Eric Juergens and William D. Regner — said many boards are asking whether artificial intelligence now requires a dedicated specialist in the boardroom. Their answer was more cautious than that label suggests. (corpgov.law.harvard.edu) (debevoise.com) The memo says the pool of people with both deep AI expertise and the qualifications to serve effectively as a public-company director is limited. It also says only a small percentage of companies are so dependent on AI that they need a board seat built around that specialty. (corpgov.law.harvard.edu) (debevoise.com) Debevoise said a designated expert can change board behavior in ways directors usually try to avoid. Other directors may defer too much to one person, weakening the challenge-and-debate process boards use to test management’s plans. (corpgov.law.harvard.edu) (debevoise.com) The memo also points to conflict risks. People with deep AI backgrounds often have investments in AI companies or business ties to vendors, and those relationships can complicate independence questions. (corpgov.law.harvard.edu) (debevoise.com) Instead of treating AI as a badge one director wears, the memo says boards can rely on management and outside specialists when technical questions arise. It cites Section 141(e) of Delaware’s corporate law, which says directors are “fully protected” when they rely in good faith on selected officers, employees and experts within their competence. (debevoise.com) (corpgov.law.harvard.edu) That argument lands as investor expectations around AI governance are rising. A March 11 post on the same Harvard forum, based on Glass Lewis analysis, said just over half of S&P 100 companies disclose board-level AI oversight, and fewer than one-third disclose both oversight and a formal AI policy. (corpgov.law.harvard.edu) The Glass Lewis analysis also said 65% of U.S. investors believe all companies should clearly disclose board oversight of AI governance issues and AI ethics. It said 67% evaluate AI issues case by case rather than through fixed voting policies. (corpgov.law.harvard.edu) The Harvard post does not tell companies never to recruit directors with AI experience. It says boards should add that expertise when a candidate also brings the broader judgment, independence and board fit expected of any public-company director. (corpgov.law.harvard.edu) (debevoise.com) The thread running through both pieces is that AI is moving from a technology question to a governance one. Boards are being told to show they can oversee it, but not to confuse oversight with appointing a single oracle. (corpgov.law.harvard.edu 1) (corpgov.law.harvard.edu 2)