Ask an Ethicist columnist Vera Cherepanova addresses an issue directors are increasingly facing as cyber issues and AI land on their agendas. Board members are tasked with making decisions even if they don’t completely understand the technology they’re addressing. When and how should a board member say, “I don’t know enough?”
I am an independent director. More and more often, the issues coming to our board are highly technical — whether in AI, cyber or robotics. We are being asked to approve major strategic moves involving technology and risk that many of us do not fully understand. But no one wants to look uninformed, management is confident, the advisers are polished, and the discussion keeps moving on. Is admitting you are out of your depth still part of good fiduciary conduct? And what are the good — pain-free — ways of saying “I do not know?” — Name withheld
Your dilemma is so rich because although it is wrapped in the language of today’s technological disruptions, underneath, it touches on many aspects of boardroom character and deeper issues connected to humility vs. ego, competence vs. performative confidence, board culture vs. duty of care and the real meaning of stewardship.
Boards are now being asked to exercise judgment around technologies that are evolving faster than traditional governance structures, reporting models and oversight processes. Agentic AI, quantum computing, robotics, bioengineering and advanced cybersecurity are creating new tensions around autonomy, accountability, workforce impact, capital allocation, risk oversight and stakeholder trust. How boards govern under uncertainty and oversee emerging risks and opportunities that cut across strategy, operations, talent, disclosure and reputation are the questions widely discussed in board circles.
However, another critical question underpinning how well directors will cope with these challenges is this one: If you know you are out of your depth on a matter that is material to the company, do you have a duty to say so, slow things down and ask for help?
Let’s take a quick detour to the CEO world. In recent headlines, two well-established and by any conventional measure successful leaders cited AI as a reason for stepping down:
Doug McMillon, who led Walmart over a decade, told CNBC “I could start this next big set of transformations with AI, but I couldn’t finish.” That’s why he handed over to someone he described as faster. James Quincey, who has been with Coca-Cola since the ’90s and led it through a portfolio transformation and a pandemic, has offered similar reasoning.
Both of the CEOs framed their decision as a strategic rationale: Having looked at what the next five to 10 years require, they concluded that finishing the job needs a different kind of a leader than the one who started it.
Their stories invite honest self-reflections by directors, too: What skills, experiences and mindsets will make directors effective in boardrooms over the next five years? What does a future-ready board actually look like in practice? And will directors still have what it takes to meet their oversight duty at the end of their tenure?
Speaking of duty, in Delaware-style US corporate law, directors’ duty of care requires them to make informed decisions, so the substance of the point is sound. In the UK, directors have a statutory duty to exercise reasonable care, skill and diligence.
John Weinberg, the man who laid down the foundational philosophy on governance and director qualifications in the United States, wrote in his famous 1948 Princeton thesis: “A director must not only be willing to direct, but more important … must know enough to direct,” and “the primary step to be taken is a comprehensive educational program.”
That means to exercise the oversight duty, directors need to make sure they are learning, and when they are expected to project confidence rather than acknowledge limits, that’s a cause for a serious governance concern. To reframe, the duty of care includes knowing when you do not know enough, asking for clarification, expert input or more time and should be viewed not as a weakness but as a strength.
Are you able to recognize and declare when you are out of your depth? If so, congratulations. Your board has done the work to build and reinforce high levels of honesty, authenticity and trust among fellow board members, allowing them to leave armor at the door. And no need to make it personal or dramatic. Something like, “I do not think the board yet has enough information to exercise informed judgment,” “Before we approve this, I would like an independent briefing on the risks, assumptions and downsides,” or, “Can management explain how this will affect X, Y and Z?,” help find confidence without emotional charge. Seeking guidance and support from other directors, the chair or trusted mentors might be also warranted in some cases.
Of course, the need for boards to have technical knowledge varies by company, and given that only 5%-6% of companies can be currently described as leaders in AI adoption, I do not expect many board directors yet to face their Kodak moment and follow the footsteps of McMillon or Quincey. However, false confidence is something we all need to say a hard no to from the outset. Directors on boards want to do their best and what’s right as a fiduciary duty to the communities they were drawn from, and we shouldn’t settle for anything less.
How You Handle AI Agents Says More Than You Might Think About Your Company’s Values
Are you really ready to deploy a virtual workforce?
Read moreDetailsReaders respond
The previous question came from a board director at a large listed company considering management’s proposal to introduce AI agents into customer workflows as a “digital workforce.” The dilemma revolved around whether this was merely a technology upgrade or a broader governance decision, raising questions about AI autonomy, customer experience, human labor, escalation protocols and what must remain under human judgment.
In my response, I noted: “Your question is an excellent one because management is presenting this as a technology upgrade, while the board should recognize it as something much larger. Managing a workforce that includes both humans and digital workers cannot be confined to a technology implementation project. It is a board-level governance issue for several reasons.
“The first reason is because AI is moving here from an assisting role — helping humans with certain tasks — to acting on behalf of them and, in some cases, on behalf of the company as a whole. That calls for a different oversight model, one capable of distinguishing between different levels of AI autonomy and adjusting scrutiny accordingly. Ceteris paribus, the more independently these systems act, the more closely they should be governed.
“Second, as your dilemma suggests, the actions of intelligent agents will directly inform the customer’s experience of the organization. Therefore, a useful question for the board to ask, in full honesty, is whether these agents are deployed to serve customers better or to make it harder for customers to reach the company. The commercial model provides an additional insight: If the company pays for ‘resolved’ outcomes, the system may be tuned to classify more issues as resolved regardless of whether the customer would agree.
“Third, introducing digital labor will affect human labor in ways management might be understating…” Read the full column here.
This is exactly the kind of boardroom conversation that needs to happen more. I run a restaurant company, and we had a similar moment — the “digital workforce” pitch sounds incredible until you start asking the harder questions. What happens when the AI agent makes a decision that affects a guest’s experience and nobody can explain why? Who’s accountable when the system learns a pattern that inadvertently discriminates? The companies that are going to build lasting trust with AI aren’t the ones who move fastest … They’re the ones who build the governance framework first and then move fast within it. Values aren’t a constraint on AI adoption — they’re the foundation that makes sustainable adoption possible. — SF
Vera Cherepanova
