The Director's Dilemma

AI Oversight From the Board Seat

For boards, the choice is no longer whether to engage with artificial intelligence, but how. Inaction is itself a decision, and a risky one.

In Brief

The Problem. AI is now a permanent part of healthcare and life sciences organizations, and ignoring it, delegating it entirely, or misunderstanding it is a failure of governance.

Why It Persists. Boards treat AI as a purchase, approve strategy without testing implementation, lack the shared literacy to question management, and judge value with measures built for equipment.

What Boards Should Do. Govern AI as a capability, govern the how as well as the why, raise the board's own capability, and ask for measures that fit the asset. Then record what the board asked and required.

After the cycles of excitement have peaked and receded, a simple, hard reality remains for the corporate director. Artificial intelligence (AI) is now a permanent and powerful part of how healthcare and life sciences organizations deliver care, discover products, and run their operations. To ignore it, to delegate it entirely, or to misunderstand it is not a technical oversight. It is a failure of governance.

This series has examined that duty from several angles. Governing the Gaps showed how AI tests every duty the board already holds. A Director's Heat Map showed where to spend the board's limited attention, and An Un-United Nations of AI set out the rules that apply in each jurisdiction, and the dates. They point to one mandate, which I distill into four principles. My closing argument is that the principles matter less than the record that shows the board applied them.

First, Govern AI as a Capability, Not a Purchase

AI is not an information technology project to be managed; it is a corporate capability to be governed. Unlike a new software suite, it changes after approval: models drift, vendors update them, and the populations they serve change. Governing AI therefore requires a shift from procurement to stewardship.

Boards must stop asking only about upfront cost and ask about the whole life of the system: its training, its maintenance, its potential for drift, and its eventual retirement. Stewardship also means understanding provenance. Whether built in-house or bought, a model brings its own data, assumptions, and limits, and it can carry inherited risks or biases. The board should know what the vendor terms say about training on the organization's data, notice of changes, audit rights, and indemnification. When the Black Box Breaks shows what happens when a model meets data it was never built for.

Second, Govern the "How," Not Only the "Why"

The most compelling strategic rationale for an AI initiative is worth little if the implementation is flawed. Failure lives in the details: in poor data, in workflows that do not fit, and in the reluctance of the people meant to use the tool. In August 2025, researchers at the NANDA initiative of the Massachusetts Institute of Technology reported that about 95 percent of organizations in their study had gotten zero return from their AI efforts, and that only about five percent of custom enterprise tools reached production (Claburn, 2025).

Boards must develop a healthy skepticism toward polished demonstrations and demand proof of operational readiness. That means asking hard questions about data integrity, change management, and the gated process of moving AI from a promising prototype to a dependable, enterprise-scale system. From Sandbox to Scale describes the six gates of the Avancer AI Adoption at Scale Framework and the conditions a board should attach at the last one. A board's duty is not to be impressed by the promise of a technology. It is to be convinced by the soundness of the plan to deliver it.

Third, Raise the Board's Own Capability

The tech intimidation that silences many boardrooms is a self-inflicted wound. A board that lacks the literacy to challenge management on AI is failing in its duty of scrutiny. In 2025, the National Association of Corporate Directors found that only 18 percent of public company boards had a director education plan for AI (National Association of Corporate Directors, 2025).

The solution is not to recruit a single AI expert, which concentrates understanding in one seat. It is a commitment to a baseline of shared education and the active recruitment of directors with experience in large-scale technology change. The most valuable experience is often not that of a programmer but of an executive who has overseen a difficult integration and learned what can go wrong.

Overcoming Tech Intimidation sets out how a chair can make basic questions safe to ask. Orchestrating Oversight describes how to divide the work among committees. The same discipline applies to the people the board supervises: The Augmented Executive explains why the evaluation of each officer should now include how that officer uses, monitors, and reports on AI.

Fourth, Ask for Measures That Fit the Asset

Measures designed for an industrial economy do not fit a transformative technology. A narrow focus on short-term return biases investment toward small, cost-saving applications at the expense of bolder, strategic initiatives. Boards should ask instead for sector-specific measures, such as readmission rates, trial timelines, and denial rates, paired with a strategic scorecard that assesses competitive position, the creation of proprietary data assets, and long-term resilience. Most important, the board should agree before approval how and when results will be reported. Measuring the Ghost describes how.

The Objection: No Court Has Yet Ruled on AI Oversight

The strongest objection to this series is that it runs ahead of the law. Commentators who argue that AI oversight is a fiduciary duty cite no court decision on a board's oversight of AI; they draw on oversight cases about food safety and aviation safety and treat AI as falling within them (Meson, 2026). A skeptical director might conclude that the duty is speculative.

The answer is that the duty does not depend on the technology. The test comes from In re Caremark International Inc. Derivative Litigation, 698 A.2d 959 (Del. Ch. 1996): did the board make a good-faith effort to ensure that a reporting system exists? Marchand v. Barnhill, 212 A.3d 805 (Del. 2019), applied that test with particular force to "mission critical" risks (Zauderer et al., 2023). Since January 2023, the duty also reaches officers within their areas of responsibility (Raymond et al., 2023). In September 2025, the Court of Chancery applied it to a drug maker whose board allegedly had no committee responsible for regulatory compliance (Bonadurer, 2026). When AI supports clinical decisions or regulatory submissions, it sits inside exactly those mission-critical risks. As Meson puts it, drawing on a paper by Pierluigi Matera, AI "changes the evidentiary terrain on which good faith is shown" (Meson, 2026). Waiting for the first AI case means waiting to be the defendant in it.

The Record Is the Standard

These principles are not a checklist. They are the long-standing duties of a director, applied to a new kind of asset. The standard is procedural, and the minutes are the evidence.

Consider a board that can show what it asked, what conditions it attached, and what it monitored. That board has met its duty, whatever the technology turns out to do. I have come to believe that this is the most practical advice in the series. Write it down.

In the end, the challenge AI poses to the boardroom is not about technology at all. It is about leadership: the judgment to see a fundamental shift for what it is, and the courage to adapt the lasting principles of good governance to meet it. For every director, the choice is no longer whether to engage with AI, but how.

For the Board

• Does our board treat each material AI deployment as a governed capability, with an owner, a monitoring plan, and a retirement plan?
• Do we know the provenance of each material model, including what our vendors trained it on and what our contracts allow?
• What evidence of operational readiness do we require before approving scale?
• What is our plan to raise every director's AI literacy this year?
• Which measures will tell us whether each AI investment is working, and when will we see them?
• Would our minutes show a court what we asked and what we required?

To work through these principles with your board in a Board Readiness Workshop, book a conversation with Avancer.

References

Bonadurer, J. (2026, February). Delaware court denies dismissal of Caremark oversight claims in Teligent case, highlighting board and officer compliance obligations. Jenner and Block. https://www.jenner.com/en/news-insights/publications/delaware-court-denies-dismissal-of-caremark-oversight-claims-in-teligent-case-highlighting-board-and-officer-compliance-obligations

Claburn, T. (2025, August 18). GenAI FOMO has spurred businesses to light nearly $40 billion on fire. The Register. https://www.theregister.com/2025/08/18/generative_ai_zero_return_95_percent

Meson, P. (2026, June). Guest post: AI governance is a fiduciary duty. The D&O Diary. https://www.dandodiary.com/2026/06/articles/artificial-intelligence/guest-post-ai-governance-is-a-fiduciary-duty/

National Association of Corporate Directors. (2025). AI oversight activities conducted by the board [Data table]. 2025 Public Company Board Practices and Oversight Survey. Retrieved October 21, 2026, from https://www.nacdonline.org/all-governance/governance-resources/governance-surveys/surveys-benchmarking/bpo-tables/ai-oversight-activities-conducted-by-the-board

Raymond, F. D., Lidondici, C. M., & Schiltz, T. C. (2023, February 23). Delaware Court of Chancery expands duty of oversight to corporate officers. Faegre Drinker. https://www.faegredrinker.com/en/insights/publications/2023/2/delaware-court-of-chancery-expands-duty-of-oversight-to-corporate-officers

Zauderer, F. M., Monteleone, J. P., & Fenichel, A. H. (2023, August). Guest post: Board oversight duties, recent adventures in the Delaware Chancery. The D&O Diary. https://www.dandodiary.com/2023/08/articles/director-and-officer-liability/guest-post-board-oversight-duties-recent-adventures-in-the-delaware-chancery/