Helayna challenges an old and stubborn conception of the board: a collection of accomplished people who gather periodically, review what management puts in front of them, approve the important things, question a few others, and ultimately provide their blessing to the CEO.
The proverbial rubber stamp.
The modern board should be intentional and forward-looking. It should help guide the purpose of the enterprise, challenge management’s assumptions, anticipate what is coming around the corner, and help ensure that the company creates positive outcomes while also growing revenues and shareholder value.
And lately, her thinking has me wondering about something else.
Not simply its cybersecurity.
Not simply its legal risks.
Not simply how much money it can save.
And certainly not simply how many people it might replace.
Boards should be asking a much larger question:
Because that decision is being made right now.
For the first few years of generative AI, companies could reasonably treat AI as a technology issue.
Which tools should employees use?
What information could they put into them?
How could AI improve productivity?
Agentic AI changes the stakes.
They will perform sequences of work, interact with systems, gather information, make recommendations, initiate actions, and increasingly operate alongside human talent.
That means AI implementation is becoming something much bigger than technology deployment.
It is becoming organizational design.
Every time an organization gives work to an AI agent, it implicitly decides what work remains for a human being.
Every time AI accelerates a process, someone decides what happens to the time that acceleration creates.
Every time AI provides an employee with vastly greater access to knowledge, someone decides whether that employee gains greater authority to use it.
And every time AI eliminates repetitive work, someone decides whether the worker receives more interesting work—or simply more work.
This may become one of the defining corporate governance questions of the next decade.
Processes that once took days may take hours. Tasks that took hours may take minutes. Research, analysis, administrative work and coordination may become dramatically easier.
Where does that dividend go?
The conventional answer will be obvious:
Productivity.
If ten people can accomplish what once required twelve, reduce headcount.
If a worker can complete a task 30 percent faster, give that worker 30 percent more work.
If AI increases capacity, immediately convert that capacity into output.
Some of that will happen. And some of it should.
Companies exist to produce economic value.
Boards should challenge management when efficiency becomes the entire AI strategy.
Because relentlessly converting every AI-generated efficiency into additional human workload could create extraordinarily productive companies filled with extraordinarily unhappy people.
That isn’t necessarily good governance.
And over time, it may not even be good economics.
There is another way to measure the return.
Not merely their productivity.
Their capacity.
There is an important distinction.
Productivity asks: How much more did this person produce?
Productive capacity asks: What is this person now capable of doing that they couldn’t do before?
Did AI give a junior employee access to information that once belonged mostly to senior executives?
Did it help a manager recognize a problem sooner?
Did it give someone enough time to learn a new skill?
Did it enable employees to test ideas they previously lacked the resources to explore?
Did it allow teams to spend more time with customers?
Did it make people better at exercising judgment?
Did it remove enough administrative drudgery that people could spend more of their working lives doing things that actually require human beings?
As agentic AI moves deeper into organizations, I believe directors should repeatedly challenge leadership around five things.
What work are we removing from people?
Automation should have a purpose. Boards should understand whether AI is primarily removing drudgery, eliminating jobs, augmenting expertise or fundamentally restructuring roles.
What are people gaining in return?
If employees save five hours a week through AI, what happens to those five hours? Do they gain learning, experimentation, customer interaction, collaboration or some greater measure of control over their work?
Or does the workload simply refill the space?
Is AI expanding agency or reducing it?
This may be the most important question.
AI can create workplaces in which people have extraordinary information at their fingertips and greater ability to make decisions.
It can also create workplaces in which algorithms monitor, prescribe and constrain nearly everything people do.
Those are two radically different futures of work.
Boards should know which one their company is building.
Are we making people more capable—or more dependent?
An employee who accepts every AI recommendation without understanding it isn’t necessarily becoming more valuable.
People still need judgment, curiosity, critical thinking and discernment.
The best AI environments should strengthen those human capabilities rather than quietly atrophy them.
Are people actually experiencing better work?
Employee engagement surveys won’t be enough.
Boards should want to understand whether AI is affecting workload, burnout, learning, autonomy, mobility and people’s sense that their work matters.
If management reports enormous AI productivity gains while employee exhaustion is simultaneously increasing, directors should become very curious.
Boards will naturally receive AI dashboards.
They will see capital expenditures, adoption rates, productivity improvements, security incidents, cost savings and return on investment.
Good.
They should.
It might track internal mobility.
Skills gained.
Time spent learning.
Employee autonomy.
Manager spans of control.
Workload.
Burnout.
Retention of high-performing talent.
Employee confidence using AI.
Perhaps even something as simple as the percentage of workers who believe AI has made their jobs better.
Imagine the conversation that metric could generate in a boardroom.
Management announces that its AI program generated $200 million in productivity improvements.
A director responds:
“That’s impressive. What percentage of our employees say AI improved the quality of their working lives?”
That is a very different governance conversation.
I can already hear the objection.
Corporate boards aren’t responsible for making employees happy.
Fair enough.
A board isn’t an employee wellness committee.
But happiness at work shouldn’t be dismissed as ping-pong tables, free lunches and feel-good HR programs.
Think instead about whether people experience agency, mastery, meaning and sustainable workloads.
Those things affect retention.
They affect innovation.
They affect institutional knowledge.
They affect customer relationships.
They affect whether talented people want to join an organization and whether the best people want to stay.
The interesting governance question is how often they can reinforce one another.
And perhaps this is where boards become particularly important.
Executives operate under tremendous short-term pressure.
Revenue targets must be hit.
Margins must improve.
Analysts must be satisfied.
Shareholders expect returns.
AI will present management with irresistible opportunities to generate near-term efficiencies.
Someone needs to keep asking about the enterprise five years from now.
What happens if we remove too many entry-level positions?
Where will our future senior talent come from?
What happens if people stop learning because AI performs all of the developmental work?
What happens if managers oversee algorithms rather than mentor people?
What happens if we gain efficiency while losing institutional knowledge?
What happens if our highest performers conclude that AI has simply enabled us to demand more from them?
And perhaps most importantly: What happens if we create an extraordinarily efficient company that nobody particularly wants to work for?
Corporate purpose is relatively easy to discuss when conditions are stable.
It becomes meaningful when leaders must make difficult choices.
Agentic AI will create many of them.
Companies will have extraordinary opportunities to eliminate costs, increase output and restructure their workforces.
They may decide that some portion of the AI dividend should be reinvested in human beings.
More learning.
More experimentation.
More mobility.
More decision-making authority.
More meaningful work.
Perhaps even more time.
And that brings me back to Helayna Minsk’s vision of the modern corporate board.
If boards really are evolving from passive overseers into intentional stewards of corporate purpose, then agentic AI may provide one of their greatest tests.
The question before them isn’t simply:
How much value can AI create for this company?
It is also:
What will we do with that value once it is created?
And perhaps even a little happier.
Our best days lie ahead.
Note: image in this post was rendered on ChatGPT based on our prompts.
I’m Dan Smolen. As host and executive producer of What’s Your Work Fit? I help you make your work and workplace decisions result in better and more satisfying professional experiences and outcomes. I am also a Founding Member of The Future of Work Alliance.
No posts
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.