Designing for Cognitive Sustainability in the AI Era

The overlooked risk isn’t only cognitive offloading. It’s cognitive intensification.

We’ve started to talk more and more about cognitive offloading - the way AI makes it easy to outsource thinking, and in doing so lose the mental muscle that builds expertise.

It’s a real concern, and it lands because we’ve seen this pattern before. Every major technology shift makes some tasks easier, some skills less used, and some human capabilities quietly atrophy if we don’t keep practicing them.

But I think there’s an equal and opposite risk that’s getting far less attention.

Cognitive intensification

As AI absorbs more of the workflows and frictions that sit around deep cognitive work, it doesn’t just give us time back. It changes the ratio of what fills our day. The “in-between” work that used to sit between bursts of deep thought is increasingly being compressed, automated, or removed.

That might sound like an obvious win, until you recognise what that in-between work was also doing.

For decades, busywork accidentally gave us cognitive interval training. Those pauses between demanding moments weren’t always satisfying, but they acted as a reset. You finished a hard piece of thinking, then you downshifted. You caught up on messages. You handled admin. In the middle of it, your brain recovered.

Now imagine a world where much of that coordination, drafting, summarising, scheduling, status chasing, and follow-up work is handled by AI. What happens to the rhythm of a human week when the pauses are engineered out?

What fills that space is not necessarily rest. Most likely - especially because it’s become a foundational part of the AI business case - it’s more cognitive effort, with fewer natural breaks. Not just more work, but more high-intensity work: bigger problems, faster cycles, tighter collaboration, and an expectation that we’ll use reclaimed time for “the good stuff.”

And this is where the trap sits: 50 or 60 hours a week of pure creativity and complex thinking is not a productivity dream. It’s a cognitive overload recipe. Human energy has limits. Attention has limits. The ability to collaborate intensely has limits. If you remove the natural breaks, you don’t remove the limits. You just reach them faster.

So we face a dual challenge at the same time.

• Cognitive offloading is what happens when the path of least resistance becomes the default. We hand thinking away until we no longer notice what we’re losing. We stop practising parts of our craft. We become dependent on the tool. Our judgement softens - not because we’re incapable, but because we’re out of shape.
• Cognitive intensification is what happens when AI removes support work and we don’t automatically gain balance. We gain capacity for more cognitively demanding work, and organisations often respond by raising expectations and expanding the load. The week becomes more mentally dense. The pauses disappear. The brain never fully comes off the gas.

And that’s happening against another cognitive load we also can’t ignore. The broader AI wave is creating cognitive strain in and of itself - uncertainty about roles and skills, anxiety about the pace of change, and a constant feeling of the need for vigilance to stay “relevant”. Even when nothing is “wrong,” there’s a cognitive cloud in the background, quietly consuming attention and shrinking the margin we used to rely on. We are all feeling it, whether we recognise or fully understand it, most of us have never had a higher cognitive load.

To be clear, I don’t frame this as a “wellbeing” issue focus – though it obviously has relevance there, but in this context as a performance issue. Long before burnout, cognitive overload shows up as decision fatigue, lower creativity, brittle collaboration, more conflict, less patience, and lower quality judgement.

The time dividend dilemma

The default human response to efficiency is to fill the space. We do it with money, and we will do it with time. If AI gives us a “time dividend,” we’ll be tempted to spend all of it immediately on more output, because the incentives and norms of modern work reward visible contribution.

For decades, we’ve used time as a proxy for value. Hours worked have stood in for commitment, contribution, even competence. But as people achieve 10x - even 100x - output by partnering effectively with AI, hours become a weaker signal. Impact rises. Time becomes less predictive.

That sounds like liberation. And it can be. But it also creates a new pressure. If time is no longer the measure, we reach for new signals. Many people will try to demonstrate value through constant visibility - responsiveness, presence, participation - even when it comes at the cost of recovery.

And that accelerates “cognitive intensification”.

So what should we do with this? The answer is not to slow down adoption. It’s to become more deliberate. To treat cognition as a finite resource and manage it intentionally, in the same way we manage budgets and capacity. Because both risks can be true at once.

We need to be conscious about what we offload so we don’t erode expertise.

And we need to be conscious about how we reinvest the time AI returns so we don’t overload the human system.

That means one simple shift in mindset: time saved is not automatically time gained. Time saved is time reallocated. The key question is what it gets reallocated into.

If we automate the breaks out of work, we have to design breaks back in.

Not as a wellness perk, but as a performance requirement. Breaks are not a sign of lower ambition. They’re the mechanism that sustains high-quality thinking. They’re what makes deep work repeatable - not just possible.

The organisations that win in this next phase won’t be the ones that squeeze every reclaimed minute into more output. They’ll be the ones that build a new operating model around cognitive sustainability - deciding deliberately how much of the AI dividend to cash out into recovery, how much to reinvest into capability and learning, and how much to compound into better decisions and deeper craft.

Because the future of work is not only about avoiding cognitive offloading. It’s also about surviving cognitive intensification. And that begins with noticing the drift, before the drift becomes the culture.

From personal reflection to professional responsibility

Up to this point, this has been a reflection every one of us needs. We all need to become more conscious about how we’re shifting, evolving, and maturing the way we work in this new era - what we offload, what we intensify, and where we allow the defaults to take over.

But there’s a second step. Once you can see these patterns in yourself and your team, you start to see them everywhere. You begin to notice what helps cognition and what drains it - what creates clarity, what creates fatigue, and what quietly taxes attention.

And that’s where this becomes professionally relevant for those of us who enable modern work. Because the way we design tools, workflows, and environments doesn’t just support collaboration. It shapes the cognitive experience of collaboration. It influences whether people leave a session energised and aligned, or depleted and starting over.

So if cognitive offloading and cognitive intensification are becoming defining risks of the AI era, then helping organisations navigate them is not only a personal discipline. It’s becoming part of the job.

Where this becomes real for AV - and for every space where people think together

As cognitive work becomes denser, the environments where it happens become higher stakes. Meeting rooms, classrooms, training spaces, studios, labs - any place people gather to think, learn, decide, create, and align - these are not neutral containers. They shape cognitive effort: how quickly it ramps up, how long it sustains, how inclusive it is across different thinkers, and how cleanly it recovers.

In an era of cognitive intensification, the experience of those environments can no longer be neutral.

So here’s a question: What if the job of a collaboration or learning space was not only to host cognitive effort, but to help regenerate it?

This is where AV has a unique advantage. We don’t just connect people and content. We shape the sensory layer of thinking: what groups see, what they hear, what they can hold in shared attention, and how the environment responds as the purpose of the moment changes.

Think about what becomes possible when a space is designed less like “screens and microphones” and more like a cognitive instrument.

A full wall can become a shared canvas, not merely a display. When the intent is focus, the environment can reduce visual noise, tighten the frame, stabilise context, and help a group hold a single thread without constant resetting. When the intent is ideation, it can expand the field - multiple streams of thought held side-by-side, fragments made visible long enough for patterns to emerge. When the intent is convergence and decision, the space can shift again - compressing options, tracking trade-offs, making the path of reasoning visible, capturing outcomes in a way that prevents the next session from starting at zero.

The deeper opportunity is adaptiveness. If the environment can sense intent and workflow in real time - not perfectly, just enough - it can reduce the cognitive tax that normally leaks out of human gatherings: lost time in transitions, repeated context setting, friction that drains attention, and the subtle fatigue that shows up as lower participation and weaker decisions.

I keep coming back to regenerative braking in an electric vehicle. In a conventional system, braking energy is simply lost. In a regenerative system, some of that energy is captured and returned in a different form.

Human gatherings have their own version of wasted energy: attention that gets scattered, momentum that gets lost in mode shifts, insights that disappear because they were never stabilised into a shared artefact. A cognitively intelligent environment would not only help people spend energy - it would help them keep more of the value they paid for.

None of this is about adding theatre. It’s about recognising the new constraint. If AI removes the in-between work that used to provide accidental recovery, we have to become more intentional about how cognitive effort is sustained. That includes the spaces where the most expensive resource in the room is being consumed: human attention.

Because this next phase is not only about making collaboration and learning easier.

It’s about making cognitive work sustainable - and making the environments we design actively contribute to that sustainability.

Cognitive offloading is about losing capability. Cognitive intensification is about exceeding capacity. Our opportunity in AV is to design environments that protect both - by reducing cognitive tax and increasing the yield of cognitive effort.