75% of UK businesses are now using AI but are we heading for an employee AI brain fry epidemic?

Recent research found that three quarters of UK businesses are now using AI, with this rising up to 90% in larger corporations with up to 500 members of staff. Natalie Mackenzie, a Cognitive Strategist for business owners and leaders, explains why this could have a bigger impact on your (and your teams) brains than you realise…

For many business owners, actually hourly usage of AI is now a fully embedded part of our routines. And we are only just scratching the surface it seems. Cue new articles, podcasts and books on the new age of AI, the impact on our productivity, the possible job losses and even questions over my 12 and 10 year old children’s job prospects.

How does it show up each day in the workplace? Perhaps an employee is mid-sentence, reaching for a word  and before they even sit with the discomfort, they’ve opened ChatGPT. Or perhaps you’ve handed your weekly email marketing to an AI assistant, let it plan your schedule and summarise a document you could have read yourself. The relief is instant and convenience is undeniable for many SMEs.

Yet with the hours saved and tasks ticked off, we’re not talking about the other side of the coin.   A recent study from Boston Consulting Group found that workers who frequently use AI reported symptoms like mental fatigue, overwhelm, fogginess and struggling with decisions. In other words: AI brain fry.

So what is actually happening to your employees brain?

We don’t yet have robust longitudinal data specifically on AI use and cognitive decline but we have decades of solid research on a closely related phenomenon called cognitive offloading: the tendency to use external tools to reduce the mental effort required for a task.

To-do lists, using GPS, keeping a calendar, all of these are forms of offloading, and they have genuinely extended human capability. The concern with modern AI isn’t that we’re using tools. It’s the scale of what we’re now offloading, and how quickly.

Cognitive offloading, used deliberately, is a form of scaffolding. You use a structure,  a decision framework, a reflection habit, a written checklist, an AI tool now to support a skill while it is being built or rebuilt. Through repetition, the underlying capacity strengthens. It becomes internalised and eventually automatic. At that point the scaffold can come away, because the brain no longer needs the external support. This is how skill acquisition actually works.

Many of the people I work with are managing genuine cognitive overload and using AI to handle routine correspondence, organise information, or reduce the administrative noise around a problem can genuinely free up working memory for the thinking that matters. That is scaffolding working as intended: clearing the site so the real work can happen.

But this only holds if the scaffolding eventually comes down. And this is where I see the pattern break: people begin offloading to manage overload, find relief immediately, and gradually extend the habit as a permanent replacement. AI is always available and always obliging, there is no natural pressure to remove it. The poles and boards stay up indefinitely. And the building never learns to stand alone.

The brain is an adaptive organ. It strengthens the neural pathways we exercise repeatedly and gradually down-regulates those we rely on less. The prefrontal cortex, the home of planning, working memory, decision-making, and evaluative reasoning  is particularly responsive to this use-it-or-lose-it idea.

Equally important, and less often discussed, is the role of the anterior cingulate cortex — the ACC. This region sits at the interface of cognition and motivation, and it performs a specific and vital function: it detects when a task is difficult and signals the brain to stay engaged.

The ACC is activated specifically by cognitive challenge, by the friction of encountering a problem that doesn’t have an immediate answer. When we habitually remove the friction before the ACC has even had a chance to fire and reach for AI at the first sign of difficulty, we are chronically under-stimulating the very system responsible for keeping us productively engaged with hard things. Over time, the threshold shifts and difficulty feels harder sooner. The urge to offload arrives earlier and earlier. What we’re experiencing as ‘brain fog’ or cognitive fatigue in our businesses may in part be an ACC that has simply stopped expecting to be needed.

AI doesn’t exist in isolation. It’s layered on top of a digital environment already engineered to fragment attention. Notifications, social media, and rapid context-switching have been steadily training our brains to expect novelty every few seconds. AI, at its worst, becomes the system that removes the need to sustain cognitive effort at all.

But sustained, uninterrupted attention is precisely what deep reasoning requires. Without it, ideas are rarely processed deeply enough to be consolidated into long-term memory. Information passes through without sticking and decisions are made on the surface rather than thought through.

When my clients tell me they feel mentally drained after a day of AI-assisted work, the explanation is that the constant switching between prompts, outputs, and decisions, at the pace AI operates is cognitively exhausting. The brain is tired.

So how can we tackle the AI brain fog epidemic?

1.  Attempt it yourself first.  Before opening any AI tool, spend at least five minutes with the business problem on your own. Sketch ideas on paper, draft a rough outline, sit with the question. This ‘struggle’ activates the ACC, fires up the prefrontal networks, and drives the memory consolidation.

2.  Write before you prompt. This is particularly important for decision-making and communication at work. Write your own rough version before asking AI to help you refine it. Your draft activates the cognitive architecture of the task. AI can then support your thinking rather than replace it, providing a better output. Encourage this across your staff.

3.  Audit and categorise offloading . Keep a log for one week of everything you delegate to AI. Then sort it into two columns. The first: strategic offloading — tasks where AI reduces time so your thinking can go deeper. Routine scheduling, formatting, initial research, summarising meeting notes. This is scaffolding working as intended. The second: substitution — judgements, strategic decisions, communications that carry your voice and expertise, and that requires your specific knowledge. Most people find the second column is larger than they expected.

4.  Schedule genuine cognitive downtime: a walk without headphones, a commute without your phone, ten minutes with a notebook and no agenda. It is some of the most important cognitive work you can do.

5.  Use AI for debate.  There’s a meaningful difference between asking AI for an answer and using it to interrogate your own thinking. Ask it to challenge your reasoning, provide the opposing view, identify the weaknesses in your argument. Used this way, AI can activate rather than bypass the ACC as you’re still doing the effortful work of evaluation and judgement. The goal is a better version of your thinking.

6.  Sleep and move (the non-negotiables)  No cognitive strategy outperforms adequate sleep and regular aerobic exercise. Both are essential for neuroplasticity, memory consolidation, and the prefrontal and ACC function that underpins good decision-making. If your thinking feels foggy and slow, the first question is whether you’re sleeping well and moving enough. For the entrepreneurs I work with, these are almost always the biggest levers and the most consistently neglected.

The entrepreneurs I work with understand that their brain is their most important business asset, the foundation of every decision, every strategy, every long-term play. My work with them is not about wellness routines or productivity hacks looks but how their specific brain functions, building habits that protect and strengthen it now, and ensuring it remains sharp and resilient for decades to come.

AI can absolutely be part of that. Used as genuine scaffolding, deliberately, temporarily, with an eye on what it’s building rather than what it’s replacing. But if the poles and boards never come down, it isn’t scaffolding anymore. It’s a crutch. And the brain, like any system that stops being challenged, will quietly accommodate itself to doing less.

The habits you practise most consistently are the ones your brain strengthens. Choose them wisely.