A recent poll of 500 global CEOs found that 94% believe AI could offer better counsel than at least one of their board members. And in October 2025, Kazakhstan’s national wealth fund appointed “SKAI,” an AI system, as a voting director. Such developments signal a shift in corporate governance as large language models (LLMs) prove capable of delivering informed, data-rich advice—often beyond what overextended, part-time directors can provide. To explore this potential, researchers at Wharton’s Mack Institute and INSEAD’s Center for Corporate Governance ran an experiment comparing human and AI “boards” deliberating the same business case. The AI board, built as a multi-agent simulation, processed materials instantly, followed standard board protocols, and interacted autonomously through a structured memory system. Evaluators rated its performance against eight governance criteria. Results were striking: the AI board significantly outperformed human groups on decision quality, evidence use, inclusivity, and implementation planning. Human boards lost focus, circled decisions, and overlooked data. Experts noted that while AI boards lacked interpersonal nuance—trust-building and empathy—they excelled in structure, participation, and clarity. The authors conclude that AI boards won’t replace humans but can already serve as powerful planning tools—simulating discussions, stress-testing options, and revealing blind spots. As CEOs face pressure to show AI progress, and many question their boards’ strategic value, companies that integrate AI into governance will outpace those that don’t.
