AI Has Already Changed Our Sector. We Just Haven't Changed Our Institutions Yet.

The Debate Has Moved On

Almost exactly a year ago, I wrote about artificial intelligence and what it might mean for museums, attractions, event venues and cultural organisations.

At the time, the debate was dominated by predictions.

Would AI replace jobs?

Would visitor experiences become automated?

Would organisations become smaller, faster and more efficient?

Twelve months later, I think we can answer at least part of that question.

The disruption has arrived.

The interesting thing is that it hasn't arrived where most people expected.

The robots haven't taken over our museums. Visitors are not being welcomed by holograms. Event campuses still need managers, technicians, security teams, cleaners, caterers and operational leaders. Buildings still leak. Contractors still miss deadlines. Visitors still complain about queues and parking.

The fundamentals of running places remain remarkably unchanged.

Yet something important has shifted beneath the surface.

The Real Disruption Is Acceleration

Over the past year, I have watched senior teams move at a pace that would have seemed extraordinary only a few years ago. A piece of analysis that once took several days can now be produced in an afternoon. Research that might previously have involved multiple colleagues and several meetings can often be assembled within hours. Information that was once difficult to access has become almost instantly available.

Not because AI is making decisions.

But because it is removing friction.

I was reminded of this recently during a discussion about a complex operational issue. It involved historical information, multiple policy documents, operational data and several possible courses of action. A few years ago, gathering the relevant information would have taken a week. Papers would have been commissioned. Meetings arranged. Drafts circulated. Comments reviewed.

Instead, within a matter of hours, the Op’s SLT team had assembled the information, tested the assumptions and moved towards a decision.

The quality of the judgement remained human.

The speed was entirely different.

That is where I think many commentators have misunderstood what is happening. The real disruption is not automation. It is acceleration.

More Productive, But Not Less Busy

The challenge is that acceleration creates its own consequences.

For years, time acted as a natural constraint within organisations. There were only so many reports that could be written. Only so many presentations that could be prepared. Only so many funding applications that could be submitted. Capacity was limited by the practical reality of how long work took to complete.

AI is steadily removing some of those constraints.

At first glance, that sounds entirely positive. Research from McKinsey and others suggests workers are adopting AI faster than many leaders realise and are reporting significant gains in productivity. The evidence increasingly suggests that many routine tasks can be completed more quickly and more efficiently than before. That matters in a cultural sector where resources remain stretched and expectations continue to rise.

But there is another interpretation.

What if productivity is not the same thing as progress?

History gives us plenty of reasons to be cautious. Whenever technology increases productivity, organisations rarely bank the saving. More often, they increase expectations.

The report that once took two days now takes four hours.

So instead of producing one report, we produce three.

The presentation that previously took a week can now be assembled in a day.

So instead of creating one presentation, we create several.

The efficiency gain disappears into the system.

“Many leaders now find themselves in a curious position. They are undoubtedly more productive than they were a year ago. Yet many would struggle to say they feel less busy. If anything, the opposite appears true.”

The Burnout Question Nobody Is Asking

This may be the most under-discussed aspect of AI adoption.

The technology is creating capacity faster than organisations are learning how to manage it.

The result is that some senior teams are moving faster than ever whilst simultaneously edging closer to burnout.

Every efficiency gain creates an expectation of greater output.

Every time saving becomes an opportunity to add another task.

Every shortcut becomes a reason to increase expectations.

Technology changes capability.

It does not automatically create capacity.

Without conscious leadership, organisations risk replacing inefficiency with overload.

The Funding Model Challenge

This becomes particularly important when viewed against the backdrop of the wider funding environment.

Across the cultural and heritage sector, organisations continue to face difficult choices. Public funding remains under pressure. Costs continue to rise. Commercial income has become increasingly important. Competition for grants and philanthropic support remains intense.

In that environment, productivity gains are understandably attractive.

Yet they also raise uncomfortable questions.

Many funding models are still built upon assumptions developed before widespread AI adoption. Grant applications often assess organisational capacity through staffing structures, project teams and delivery resources. Funders routinely evaluate value for money through measures that assume a particular relationship between effort, time and output.

What happens when those assumptions begin to change?

If an organisation can now produce certain outputs more efficiently than it could two years ago, should funding expectations remain the same?

If a team can deliver more with the same resources, how should success be measured?

These questions are not theoretical ,They are rapidly becoming practical.

Why Our KPIs Need Rethinking

The same challenge applies to organisational performance.

Many of our KPIs were designed for a world where information was scarce and producing outputs required significant effort.

Reports completed ,Strategies produced ,Applications submitted and Policies reviewed.

But when AI can accelerate all of those activities, output becomes a far less useful measure of performance.

The number of reports produced tells us very little.

The quality of decisions tells us much more.

Increasingly, I suspect organisations will need to focus less on activity and more on outcomes. Less on volume and more on impact. Less on production and more on judgement.

The institutions that recognise this shift early are likely to gain a significant advantage.

The Job Market Is Already Changing

The impact is also becoming visible in the labour market.

For decades, expertise created natural barriers to entry. Knowledge was difficult to access. Experience accumulated slowly. Specialist capability commanded a premium.

AI is beginning to level parts of that playing field.

A capable graduate with access to powerful tools can now undertake work that previously required years of accumulated experience. Research can be accelerated. Analysis can be enhanced. Content can be produced more quickly.

This does not remove the need for expertise.

But it changes what expertise looks like.

Increasingly, organisations will value judgement more highly than information. Critical thinking more highly than content production. Decision-making more highly than administration.

The leaders who thrive will not necessarily be those who know the most.

They will be those who ask the best questions.

Human Judgement Becomes More Valuable

And perhaps that brings us to the biggest misconception of all.

The popular narrative suggests that AI reduces the value of human expertise.

I increasingly think the opposite is true.

When information becomes abundant, judgement becomes more valuable.

When everyone has access to similar tools, the differentiator is no longer information. It is wisdom.

The ability to ask the right question ,to challenge assumptions ,the ability to interpret complexity , to build trust and the ability to lead people through uncertainty.

These are not technological capabilities.

They are deeply human ones.

Time To Catch Up ?

Which is why I find myself arriving at a conclusion that feels slightly contrary to much of the current debate.

The biggest impact of AI may not be technological at all.

It may be institutional.

The technology is moving quickly.

The people are adapting.

The job market is already responding.

Yet many of our funding models, governance structures, performance frameworks and leadership assumptions still belong to a world that no longer exists.

A year ago, we were asking whether AI would change our sector.

Today, that feels like the wrong question.

The change is already underway.

The more interesting question is whether our institutions are capable of changing quickly enough to keep up.