As artificial intelligence becomes embedded across every aspect of business, many organisations are understandably focused on the productivity gains the technology promises to deliver. Boards, too, are exploring how AI can streamline governance processes and help directors absorb information more efficiently. In a business environment defined by intense competition, geopolitical uncertainty and growing complexity, the appeal is obvious. Yet in my conversations with directors and executives over the past year, it has become clear that AI is forcing boards to confront a deeper issue – whether current governance models are genuinely equipped for the pace, complexity and uncertainty organisations now face. They are asking fundamental questions about how AI may reshape leadership itself – how decisions are made, how judgement is exercised, and whether boards are equipped for the pace and scale of change now taking place.
This matters because boards are already operating under considerable strain. As regulatory expectations and stakeholder scrutiny have increased, board packs have grown longer and agendas more crowded. Directors often describe evenings and weekends spent wading through hundreds of pages of pre-meeting material, only to leave the meeting feeling dissatisfied because insufficient time was spent discussing the issues most likely to shape the organisation’s future. Directors are expected to oversee an extraordinary breadth of issues, ranging from cyber risk and sustainability to workforce transformation and geopolitical instability, often with limited time and incomplete information. The result can be a form of cognitive overload in which boards become highly effective at reviewing detail, but less effective at creating space for strategic debate and meaningful challenge.
New categories of risk
AI offers an opportunity to reduce some of the administrative burden that has accumulated around governance over the past decade. But it also introduces entirely new categories of risk and complexity for boards. Unlike previous waves of enterprise technology, AI does not sit neatly within a single function – it influences hiring decisions, customer interactions, compliance and competitive strategy simultaneously, while also changing how organisations themselves process information and make decisions.
Governing AI therefore requires a different kind of conversation from the one many boards are used to having. It demands a more interdisciplinary, dynamic and forward-looking approach, as well as a greater willingness to interrogate assumptions and challenge consensus. The most effective boards are rarely those that simply move fastest. More often, they are the ones that can also sustain high-quality discussion under pressure, creating environments where alternative perspectives are surfaced early and where directors remain willing to question both the information presented to them and the reasoning behind it. This is precisely why some organisations are beginning to experiment with AI not simply as an administrative assistant, but as a strategic sparring partner for the board itself. Last month, Board Intelligence partnered with Lloyds Banking Group to explore how AI could help executives and directors interrogate assumptions, identify gaps in reasoning and surface perspectives that might otherwise be overlooked ahead of high-stakes discussions. One director described the experience to me as being less like using a chatbot and more like having access to an additional perspective in the room, one capable of challenging logic, highlighting blind spots and encouraging deeper scrutiny before decisions were made. The objective was not to replace human judgement, but to strengthen it.
Faster isn’t better
That distinction is important because much of the first generation of ‘AI for boards’ has focused primarily on precisely the opposite. While AI’s automation and summarisation capabilities undoubtedly save time, a weak board paper summarised by AI is still a weak board paper. Faster processing of information does not produce better judgement. The more exciting opportunity lies in augmentation, using AI to improve the quality of challenge, broaden perspectives and create more space for thoughtful discussion around the board table. This only works if boards resist the temptation to treat AI purely as a time-saving tool.
Governance has always depended on deeply human qualities that technology alone cannot replicate – curiosity, scepticism, integrity and the confidence to challenge consensus when necessary. If AI merely accelerates boards’ existing weaknesses, whether that is overconfidence or a reluctance to challenge prevailing views, then faster governance may simply become less effective governance delivered at greater speed.
As the outlook for growth falters and disruption intensifies, the quality of a board’s judgement will become a far greater competitive differentiator than it has been historically. The boards that benefit most from AI won’t be those that rely on it for answers, but those that use it to ask better questions, challenge assumptions and sharpen their thinking.
