The problem is that UK and European policymakers seem to have absorbed the same logic. They look at what the big American AI companies are doing and conclude that, to stay competitive, we need our own version of the same thing. More data centres. More sovereign compute. More frontier models. More public money trying to ensure we are not left permanently dependent on American infrastructure.

Some of that is understandable. Europe does need more technical capability. The UK cannot afford to be naïve about AI sovereignty. There are obvious risks in outsourcing too much of our intelligence layer to a small number of US companies, especially when the US government has already shown it is willing to restrict access to advanced chips and AI technologies when it suits its geopolitical interests. What looks like a sensible commercial dependency today can become a point of leverage tomorrow.

But I worry we are making a fairly basic strategic mistake.

We are not fast followers in this race. We are slow followers. And worse, we are slow followers with very different conditions.

The US has vast amounts of land, cheaper energy in many regions, deeper capital markets, the major cloud platforms, the frontier labs, the chip relationships and a handful of enormous technology companies willing to spend staggering amounts of money. The UK and Europe have a very different hand to play. We have more expensive energy, tighter planning constraints, less available land, shallower pools of growth capital and far fewer companies with the balance sheets of Microsoft, Amazon, Google or Meta.

Trying to beat America at America’s version of the AI race feels like a poor place to start.

So perhaps we should stop asking how we catch up, and start asking where we should deliberately take a different path.

That is why I’m becoming much more interested in open-weight models and edge computing. Not because they are a consolation prize, or a cheaper imitation of frontier AI, but because they point towards a strategy the UK and Europe could actually be good at: smaller models, running closer to the user, using less energy, protecting more private data, and doing more with less.

Doing more with less is not some romantic European virtue. It is a constraint we have had to get good at. We do not have the deep pockets of the US hyperscalers. We do not have endless land for server farms. We do not have abundant cheap energy. We do, however, have a long tradition of caring about privacy, institutions, regulation, standards, interoperability and the public consequences of technology.

Instead of treating those things as weaknesses in the AI race, perhaps we should build around them.

Most people are not trying to solve theoretical physics over breakfast. They are trying to summarise a document, search their notes, draft an email, organise a project, query a company knowledge base, write a bit of code, understand a contract, generate options, or turn a messy meeting transcript into something usable.

For that kind of work, “best model in the world” may matter less than “good enough, cheap enough, private enough and close enough to the user.”

Yet most of the current infrastructure debate assumes intelligence will keep being accessed remotely, in the same way we access storage, software, music, maps and almost everything else. You ask a question; the request disappears into somebody else’s infrastructure; the answer comes back a few seconds later.

There are good reasons for this. Training large models is expensive. The best systems still require serious compute. Most companies do not want to manage infrastructure. Consumers certainly do not. The cloud is convenient, and convenience usually wins.

But the current AI build-out also relies on a very particular assumption: that the future of intelligence will remain centralised enough to justify the cost.

That is not a given.

The railway analogy gets used a lot in discussions about AI investment. Yes, the argument goes, there may be a bubble. Yes, some investors may lose money. But the infrastructure will remain useful. The railway companies of the nineteenth century may have overbuilt, but the tracks still changed the world.

It’s a comforting analogy. I’m just not sure it works.

A railway line does not become dramatically less useful because somebody invents a better railway line three years later. The route, land, stations and physical network can hold value for decades. GPUs are different. They are extraordinary pieces of engineering, but they sit inside a brutally fast replacement cycle. Today’s frontier chip becomes tomorrow’s inference chip, then the day after tomorrow’s second-tier asset.

The land may hold value. The power connections may hold value. The cooling and networking may hold value. But the most expensive part of the build-out is not permanent in the way railway tracks are permanent. It is more like buying a warehouse full of Formula One cars and calling it transport infrastructure.

Maybe demand for AI compute will be so large that every generation of GPU finds profitable work. But we should be careful about treating compute as if it were a timeless public good. A data centre full of ageing chips is not the same thing as a bridge, a port or a railway line.

This becomes even more questionable if demand starts moving away from the centre.

Open-weight models have been improving quickly. They do not need to beat OpenAI, Anthropic or Google’s frontier models to change the economics of the market. They only need to become good enough for the fat middle of everyday work.

That is usually how technology markets get interesting. The technically superior product does not always capture the most value. Often the good-enough product wins because it is cheaper, more adaptable, easier to own, easier to modify, or simply closer to the context in which it is used.

Personal computers did not beat mainframes because they were more powerful. MP3s did not beat CDs because they sounded better. Digital cameras did not initially beat film because they produced richer images. They won because they changed the pattern of use.

The same could happen with AI.

If a law firm can run a capable model locally across its own documents, why send sensitive client material to a remote provider for every query? If a hospital can keep patient data inside its own infrastructure, why make external inference the default? If a bank wants auditability, control and predictable costs, why rely entirely on a black-box model accessed through an API? If a design team wants an assistant that understands its research archive, product decisions, brand guidelines and customer history, there is an obvious appeal in keeping that intelligence close to the work.

This is not just about cost, though cost matters. It is about privacy, latency, control, resilience and trust. It is also about fit. The more personal or organisational context an AI system needs, the stranger it feels to keep outsourcing that intelligence to a distant server.

That is where edge computing becomes interesting to me.

I’m not imagining a grand return to beige boxes and home networking projects. Most people do not want to manage infrastructure, and they certainly do not want to become part-time sysadmins just to use AI. If local intelligence is going to work, it has to disappear into the devices and environments people already use.

Your phone might run small models for quick, personal tasks. Your laptop might run larger ones for writing, analysis and creative work. Your company might run its own internal model across products, customers, policies and institutional memory. And yes, some homes may eventually have something closer to a personal intelligence server, but only if it feels more like a router or a smart speaker than a hobbyist project: quietly there, mostly invisible, useful because it coordinates the AI running across your devices.

You can imagine Apple doing something like this surprisingly well. Not another loud black box for enthusiasts, but something more like a Mac mini designed for local intelligence: a small, quiet device with a decent GPU, a private local model, and deep access to your own documents, messages, photos, calendar and notes. You would get something that feels closer to a really capable ChatGPT-style assistant, but with the context stored in your home rather than sprayed across a cloud service. For European users, that kind of model could be especially attractive. It gives you convenience without quite the same privacy trade-off. It feels like a very Apple-shaped answer to the AI problem.

That may sound slightly odd now, but so did having a media server once. The difference is that AI is more intimate than media. It wants context. It wants memory. It wants access to the messy private material that makes it useful. For some tasks, that makes the edge a more natural place for intelligence to live.

There will still be cloud models, of course. You might call out to a frontier model when the task is genuinely hard, when you need the best available reasoning, or when local models are not enough. But that does not mean every mundane act of AI-assisted work needs to be served from a hyperscale data centre.

A better pattern might be hybrid: local intelligence for the everyday layer, frontier intelligence for the exceptional layer.

This is where the policy conversation starts to feel strangely unimaginative.

If America is building centralised intelligence at massive scale, Europe could focus on distributed intelligence that is private, efficient, inspectable and close to the user.

That would mean taking open-weight models seriously as a strategic asset, not merely as a cheaper substitute for proprietary systems. It would mean investing in small and medium-sized models optimised for European languages, industries, public services and regulatory contexts. It would mean supporting inference at the edge: on devices, in offices, in hospitals, in factories, in local authorities, in schools and inside companies that cannot or should not send their most sensitive data to a remote model provider.

The prize is not national bragging rights on a benchmark table. The prize is useful AI capacity embedded throughout the economy.

There is a historical irony here. Britain had an enormous first-mover advantage in the steam age and the Industrial Revolution. We built early, scaled early, and for a time owned markets while everyone else was catching up. But early infrastructure can become a burden as well as an advantage. You inherit the old railways, the old factories, the old housing stock, the old energy systems. Newer countries can sometimes move faster precisely because they have less to rip out first.

AI infrastructure may follow a similar pattern, only on a much shorter timeline.

If the largest American companies want to spend hundreds of billions inventing the future, perhaps we should let them. Some of that work will be useful. Some of it will push the frontier forward. Some of it may create infrastructure the rest of the world can learn from. But it does not follow that the UK and Europe should rush to pour scarce capital into the same kind of infrastructure just as the underlying technology is changing so quickly.

The risk is not just that we arrive late. It is that we arrive late, spend too much, and inherit an ageing version of someone else’s architecture.

A more edge-focused strategy gives us more room to move. It does not require us to bet the farm on massive centralised compute before the economics are clear. It lets us benefit from open-weight models as they improve. It keeps more intelligence close to the data. It gives companies and public institutions a way to adopt AI without sending every sensitive query across the Atlantic. And if large parts of the current compute boom do turn out to be overbuilt, we will not have sunk quite so much national effort into infrastructure that is hard to unwind.

That feels like the more interesting bet.

Not a slightly weaker copy of the American compute race, but a more distributed, efficient and trusted model of AI adoption. One that treats privacy as a design constraint rather than a regulatory burden. One that makes open-weight models part of public and commercial infrastructure. One that helps useful intelligence live closer to the people, companies and institutions that need it.

The future of AI may still be built in giant data centres. But a surprising amount of it may end up much closer to home.

If that happens, the UK and Europe should not treat local, efficient, privacy-preserving AI as a second-best outcome. It may be the best strategy we have.

The engineer says, “But it’s obvious.”

The founder says, “But this is how the market works.”

The product manager says, “But users asked for this.”

And often, everybody is right from inside their own model of the world. The engineer understands the system. The founder understands the strategy. The PM understands the roadmap. The designer is trying to hold the experience of the person arriving cold, with none of that context, none of that history, none of that internal language, and usually very little patience.

I have a slightly half-formed theory that reading fiction as a child may make you better at this sort of work.

Not just design, although design is the obvious place to start. Product management too. Founding especially. Any role where you have to understand how different people see the same situation, then make something that works across those different models of reality.

A lot of non-fiction asks you to follow an argument. The author has a position. They gather evidence, build a case, and try to persuade you that this is the right way to see the world. At its best, this is brilliant. It can sharpen your thinking, give you language for things you half understood, or help you reason from better evidence.

But it is still usually one mind making a case.

Fiction does something different. Fiction rarely gives you one clean version of reality. It asks you to sit inside other people’s heads for a while. Often several of them. You see the same event through different characters’ eyes. One person sees danger. Another sees opportunity. Someone else misses the point entirely. They are not simply “right” or “wrong.” They are acting from their own fears, loyalties, histories, desires and blind spots.

Sometimes the narrator isn’t even reliable.

That may be one of the more useful lessons fiction teaches you: nobody is seeing the whole picture. Everybody is moving through the world with a partial model of reality, including you.

That feels very close to good product work.

Discovery only seems like a waste of time if you already believe you have the right model of the situation. If the problem is obvious, the user is obvious, the buyer is obvious, the solution is obvious, then talking to more people can look like delay dressed up as process.

But if you believe different people are seeing different parts of the truth, discovery starts to look very different.

It becomes a way of building a fuller picture.

You talk to users, buyers, support teams, salespeople, domain experts, sceptics, beginners and edge cases because each of them is holding a different fragment of reality. None of them has the whole answer. But neither do you.

This is where designers can be irritating in product discussions. They keep refusing to accept the organisation’s view of the domain as the only legitimate one. They keep saying things like, “I know that’s how we describe it internally, but would a customer use that word?” or “I understand why the flow works like this, but would somebody coming to this fresh know what to do next?”

To the rest of the team, this can sound like fussiness. Another round of naming debates. Another objection to the sensible, logical, efficient thing.

But often the designer is doing something more specific. They are trying to step outside the expert model and reconstruct the beginner’s model.

This is essentially what a cognitive walkthrough is for. You take a product you already understand and ask, step by step, whether someone else would reasonably know what to do. Not an expert. Not a teammate. Not the person who sat through three months of roadmap meetings. Someone with a goal, a bit of context, a few assumptions, and no interest in your internal taxonomy.

You’re not asking, “Can I use this?”

You’re asking, “Would someone coming to this fresh, with limited information, understand what this means, notice the right thing, and believe it will move them closer to their goal?”

That distinction sounds small, but it is often the entire job.

Teams regularly mistake their own fluency for clarity. They have spent so long inside the product that the product’s logic feels natural. They know what the button does. They know why the form has six steps. They know why the pricing page uses that language. They know what the feature is called in the database, what the sales team calls it, and why the original name was politically impossible to change.

A new user knows none of that.

They arrive with their own vocabulary, their own expectations, their own anxieties, and their own sense of what should happen next. If the product asks them to think like the company before it helps them get what they came for, the product is probably worse than the team thinks it is.

You see this in usability testing all the time. The first person struggles and someone in the room quietly blames the participant. The second person struggles and maybe the task gets blamed. By the fifth or sixth person, when the same pattern keeps appearing, the room changes. People stop laughing. The problem is no longer with the user. The problem is that the interface was designed around the team’s mental model, then mistaken for reality.

Good designers are often quicker to believe the confusion.

They are more willing to ask, “What did this person think was happening?” rather than “Why didn’t they get it?” That is a very different question. It treats user behaviour not as a failure of intelligence, but as evidence of a different model of the situation.

This is where the link to fiction feels interesting to me.

When you read a lot of fiction, you get repeated practice in seeing behaviour from the inside. Characters make choices that might seem irrational from the outside, but once you understand what they know, what they fear, what they want, and what they have misunderstood, their actions start to make sense. You learn to ask, “What would make this behaviour reasonable from their point of view?”

That is very close to abductive reasoning.

Deductive reasoning starts with rules and works out what must be true. Inductive reasoning looks at patterns and works out what is probably true. Abductive reasoning looks at a messy situation and asks what explanation would make the most sense of it.

Designers do this constantly. So do good PMs and founders.

A user ignores the obvious button. A buyer says they love the product but never converts. A customer churns after telling you the product was exactly what they needed. A founder hears ten different versions of the same problem and has to work out what is really going on underneath.

The lazy answer is to take behaviour at face value.

The better answer is to ask what model of the world would make that behaviour make sense.

Maybe the user did not understand the label. Maybe the buyer was being polite. Maybe the customer liked the product but could not persuade their boss. Maybe the market is not rejecting the product, but rejecting the way the company is framing the problem.

This is the work. Not simply collecting opinions, but reconstructing the different realities people are operating inside.

Of course, plenty of people who read fiction as children become terrible designers. Plenty of people who never cared for novels become brilliant founders, researchers, PMs or engineers. I’m not claiming some neat causal link where Jane Austen turns you into a better product thinker and business books make you worse.

But I do think fiction gives some people early practice in a skill product teams badly need: the ability to suspend your own certainty long enough to inhabit someone else’s.

That skill is harder than it sounds.

Especially in companies, where certainty is often rewarded. The person with the strongest point of view sounds decisive. The person asking for more perspectives can sound slow, vague or difficult. But certainty too early is often just a polished form of blindness. It feels efficient because it removes friction. It also removes the chance to discover that your model of the world is missing something important.

Maybe that is why designers, at their best, are such useful troublemakers.

They keep dragging other realities into the room. The confused user. The sceptical buyer. The support agent dealing with the fallout. The beginner who doesn’t know your acronyms. The customer who sees your category differently. The person for whom the “obvious” thing is not obvious at all.

That can be annoying.

It can also be the difference between building a product that makes sense to the company and one that makes sense to the people it is meant to serve.

So yes, I wonder if reading fiction as a kid makes you a better designer, product manager or founder. Not because it makes you nicer. Not because it gives you some vague empathy halo. But because it teaches you, early and repeatedly, that people do not move through the world as generic users. They move through it as particular people, with partial information, private motivations, unreliable stories and very different ideas about what is going on.

And if you can’t imagine that, you’ll keep designing for the one person whose perspective already dominates the room.

That is the problem I would ask a new prime minister to confront.

Not just how do we grow GDP? Not just how do we reduce the deficit? Not just how do we attract more investment into London, Oxford and Cambridge?

But how do we build an economy where a young person in Hastings, Stoke, Sunderland, Cardiff, Plymouth, Dundee, Hull or Middlesbrough can look around and see a plausible future?

At the moment, too many cannot.

We have created an economy with elite islands and a great deal of economic shallows. We send more and more young people to university, but too often the jobs they graduate into do not justify the promise they were sold. We celebrate startups, but many of the best opportunities cluster in London. We talk endlessly about innovation, while failing to build the industrial, technical and vocational base needed to turn ideas into durable regional prosperity.

If I were advising a new prime minister, these are the areas I would focus on.

1. Stop pretending London can carry the whole country

London is one of Britain’s greatest assets. It is a global city in a way very few places are. Finance, law, media, government, culture, venture capital, universities, tourism and international talent all compound there. It generates tax, exports, prestige and soft power.

But London’s success has become a national anaesthetic.

It allows politicians to look at aggregate numbers and pretend the economy is basically functioning. It allows Britain to appear richer, more dynamic and more globally competitive than much of the country feels. If you removed London from the national picture, the UK would look dramatically poorer.

That is not an argument for weakening London. Quite the opposite. Britain should keep investing in London’s strengths. The City may not have collapsed after Brexit, but it has lost some of its inevitability. Frankfurt, Paris, Amsterdam and Dublin have all benefited from the UK making itself less central to European finance. London remains powerful, but it is no longer quite as automatic a choice as it once was.

A serious government would do two things at once: protect London’s global role, while refusing to let it become the only serious economic engine in the country.

That means building other places with real specialisms, not just giving them better slogans.

2. Build regional specialisms, not generic “growth zones”

Every government says it wants to help the regions. The problem is that the language is usually too vague. Levelling up. Northern Powerhouse. Growth zones. Innovation clusters. Investment corridors.

Most of it sounds plausible. Too little of it survives contact with reality.

Places do not become prosperous because a minister announces a zone. They become prosperous because they develop a dense concentration of skills, employers, suppliers, institutions, investors, infrastructure and tacit knowledge around something the world actually wants.

Manchester and Salford have a credible claim around media, broadcasting, sport, digital production and parts of the creative economy. Cardiff and South Wales have television production, compound semiconductors and green industry. Bristol has aerospace, robotics, climate technology and creative technology. Sheffield has advanced materials. The West Midlands has deep manufacturing heritage. The North East has opportunities in batteries, offshore wind, rail and defence. Dundee, Leamington Spa and Brighton have strengths in games and interactive media. The South West has marine technology, food, renewables and tourism.

These are the beginnings of economic stories. They need to be treated as such.

The point is not that every town needs to become a mini-London. That is impossible and undesirable. The point is that every region needs a small number of credible specialisms it can build around.

That requires uncomfortable focus. Government cannot sprinkle money evenly and expect transformation. It has to make bets. It has to say: this place has a real chance to be nationally or internationally important in this field, and we are going to back it consistently for twenty years.

Not one parliament. Not one press release. Twenty years.

3. Rebuild the missing middle of the economy

Britain has become strangely obsessed with two kinds of company: very large incumbents and venture-backed tech startups.

Both matter. But neither is enough.

What we are missing is the layer of specialist, skilled, export-capable firms that sit between the corner shop and the global giant. Germany has its Mittelstand: often family-owned, regionally rooted, technically excellent businesses that dominate narrow global niches. They make components, tools, machines, instruments and systems most consumers never think about, but which the world depends on.

Britain has versions of this. Advanced motorsport engineering. Aerospace. Precision manufacturing. Medical devices. Marine engineering. Defence. Scientific instruments. Games. Specialist materials. But the layer is too thin.

This is where industrial policy should focus.

Not just on the next AI unicorn. Not just on software businesses that can raise a big seed round and hire in London. We should be backing companies that make difficult things well: clean energy components, robotics, medical equipment, lab hardware, advanced materials, retrofit systems, modular housing parts, marine technology, precision tools, nuclear components, battery systems and high-quality industrial machinery.

These businesses may never produce the valuations of software companies. But they create something Britain badly needs: skilled jobs, apprenticeships, exports, local supply chains, and a reason for young people to stay in or return to their region.

A country cannot live on apps, coffee shops and consulting decks alone. It needs to make things too.

4. Treat vocational education as elite infrastructure

One of Britain’s great mistakes has been to treat vocational education as second best.

For decades, the social bargain has been: go to university, get a degree, become more employable. That has worked for some. But it has also created a large number of smart graduates with expensive credentials and no clear career path.

At the same time, we have shortages of technicians, machinists, welders, electricians, fabricators, lab technicians, heat-pump installers, retrofit specialists, CNC operators, nuclear engineers, battery specialists, robotics operators and people who can bridge design, engineering and production.

This is absurd.

If we want a more productive economy, we need to stop acting as if the only respectable route into adulthood is a three-year degree followed by laptop work.

A prime minister should make technical education a national priority. Not as a consolation prize for people who “aren’t academic”, but as a high-status route into serious, well-paid, skilled work.

That means better further education colleges. Stronger apprenticeships. More employer-linked training. More applied technical universities. More modular adult learning. More investment in workshops, labs and equipment. And, crucially, it means building the employers that can absorb those skills.

Training people for jobs that do not exist is just a more elaborate form of disappointment.

5. Make universities engines of local productive capacity

Britain’s universities are genuinely world-class. They attract international students, produce research, generate soft power and support local economies. But we should be honest about the limits of the current model.

Too many universities are trying to behave like research-intensive institutions. Too many degrees offer weak labour-market returns. Too many graduates end up underemployed. Too much research fails to turn into domestic industrial capability. Too much of the value created by educating overseas students leaves the country when they do.

The answer is not to weaken universities. It is to connect them more deliberately to the productive economy.

Every major university should be asked a simple question: what is your role in the economy around you?

Not just how many students do you teach? Not just how many papers do you publish? But what industries are you helping to build? What employers are you working with? What technical skills are you producing? What spinouts are staying local? What supply chains are forming around you? What regional advantage are you deepening?

A strong regional economy might need one research university, one applied technical university, several excellent FE colleges, employer-led apprenticeship networks, shared labs, prototyping facilities, local R&D centres and patient regional capital.

At the moment, too many places have the education layer without the employer layer.

That has to change.

6. Repair the high street as economic infrastructure

The high street is often discussed sentimentally, as if the question is whether we can bring back the butcher, the baker and the old department store.

That misses the point.

The high street is not just retail. It is civic infrastructure. It is where local economies become visible. It is where people meet, eat, work, browse, repair, learn, organise, start things and feel whether their town is moving forwards or backwards.

When the high street dies, the damage is not just commercial. It is psychological.

The current tax system does not help. Business rates punish visible, place-based businesses in a way that many online businesses avoid. A shop, café, studio, small restaurant or workshop carries fixed costs that an online platform does not face in the same way. Then we wonder why town centres hollow out.

A serious government would rebalance this.

Business rates should be reformed to reduce the burden on small, independent, place-based businesses, especially in struggling towns. New businesses should receive tapered support in their first few years, when fixed costs are most dangerous. Empty properties should be brought back into use. Landlords should not be allowed to sit indefinitely on dead frontage while waiting for unrealistic rents.

But renewal cannot mean nostalgia. The high street of the future will not be purely retail. It should be mixed-use: food, childcare, clinics, workspaces, libraries, markets, evening economy, training spaces, repair shops, studios, small-scale production and housing above shops.

The goal is not to recreate 1987. It is to make town centres useful again.

7. Take coastal towns seriously

Some of the most politically alienated places in Britain are not large post-industrial cities. They are coastal towns.

Many seaside towns were built around a domestic tourism model that no longer works at scale. Cheap flights and package holidays pulled away the middle-class visitor base. What remained in many places was seasonal work, poor housing, weak transport, fragile public services, ageing populations and limited career options.

You can see the consequences in places like Hastings, Blackpool, Great Yarmouth, Clacton and parts of the Kent and Sussex coast. Some towns have managed partial reinvention through food, culture, remote workers, art, independent retail and better public realm. But many still feel cut off from the national economy.

These places need specific strategies, not generic regeneration language.

For some, the answer will be tourism, culture and food. For others, marine industries, offshore wind, port logistics, further education, care, health, creative industries or remote-work relocation. But the starting point has to be a serious question: what is this town for now?

When a place loses its economic purpose, politics curdles. People become angry not because they are irrational, but because they can see that the country has no real plan for them.

8. Fix the VAT cliff edge without punishing microbusinesses

The VAT threshold is a classic example of a policy that looks technical but shapes behaviour.

Once a business crosses the threshold, it faces a sudden jump in admin, pricing complexity and often a real hit to competitiveness, especially if it sells to consumers who cannot reclaim VAT. So some small businesses deliberately stay below the line. They turn down work. They reduce hours. They avoid hiring. They remain smaller than they might otherwise be.

That is bad policy.

But the answer is not simply to force every tiny business into VAT from day one. That would create more admin, higher prices and another reason not to start.

The better answer is to smooth the cliff edge. Create a taper. Simplify reporting. Make the flat-rate scheme genuinely useful. Treat consumer-facing microbusinesses differently from B2B consultancies. Avoid punishing the exact moment when someone tries to turn self-employment into a proper business.

More broadly, tax policy should ask a simple question: are we making it easier or harder for people to build productive enterprises?

Too often, the answer is harder.

9. Rebalance tax away from work, shops and mobility

The British tax system has accumulated too many distortions.

We tax work heavily. We tax employment through National Insurance. We tax moving house through stamp duty. We tax physical businesses through business rates. We under-tax some forms of property wealth and economic rent. We create thresholds and cliffs that encourage people and firms to behave unnaturally.

A new prime minister should not start with the fantasy that taxes can simply be cut. The demands on the state are too large: health, social care, defence, education, infrastructure, energy and an ageing population all require money.

The better question is what kind of taxes do the least damage.

That probably means lower taxes on work, hiring, mobility and small business formation. It means a more serious approach to land and property taxation. It means reforming council tax. It means reducing or replacing stamp duty. It means thinking harder about how online commerce contributes to the local infrastructure it depends on. And it means designing the system around productive behaviour rather than accidental loopholes.

Tax is not just how the state raises money. It is a set of instructions.

At the moment, too many of those instructions say: do not move, do not hire, do not grow, do not open a shop, do not cross the threshold.

10. Get closer to Europe without relitigating the referendum forever

Brexit is not the only reason Britain is struggling. But it made many of our structural weaknesses worse.

A smaller country on the edge of Europe needs trade. It needs openness. It needs access to markets, supply chains, talent and investment. Leaving the European Union made that harder, especially for goods, manufacturing and supply-chain-intensive industries.

The political class has spent too long being afraid of saying this plainly.

A new prime minister does not need to reopen every argument from 2016. But they do need to be honest that Britain’s economic future depends on a closer, more practical relationship with Europe.

That means reducing friction for goods. Deepening cooperation on energy. Making it easier for professionals, researchers, creatives and young people to move. Rebuilding scientific and industrial collaboration. And being willing to accept that access usually comes with obligations.

The fantasy version of sovereignty says Britain should make every rule alone. The useful version says Britain should have enough influence, wealth and capability to shape its own future.

Those are not the same thing.

11. Use clean energy to build industry, not just lower bills

The energy transition is often framed as a cost. It should also be framed as an industrial opportunity.

Britain is investing heavily in offshore wind, grid infrastructure, nuclear, hydrogen, storage, retrofitting and clean power. But if the turbines, components, engineering expertise and high-value supply chains are mostly imported, we will have missed a huge opportunity.

The question should not only be: how do we decarbonise?

It should be: how do we capture more of the value?

The Rolls-Royce small modular reactor programme points in the right direction. But the test is whether it creates deep domestic capability, or whether Britain ends up as a buyer, site operator and project sponsor while too much of the high-value work happens elsewhere.

The same applies to wind, housing retrofit, grid upgrades and clean industrial technology. Public investment should be used to build domestic skills, suppliers and exportable expertise.

Net zero should not just be an environmental policy. It should be one of the foundations of Britain’s next industrial strategy.

12. Build a state that can actually deliver

All of this depends on state capacity.

Britain has become very good at reviews, consultations, strategies, pilots, announcements and relaunches. It is much less good at sustained delivery.

A serious prime minister would need to rebuild the machinery of implementation. That means fewer gimmicks, clearer priorities, better procurement, more commercial skill inside government, stronger local institutions, longer funding cycles, more devolution, and a Treasury willing to distinguish investment from day-to-day spending.

One reason Britain underperforms is that nobody believes the plan will survive the next reshuffle.

Industrial strategy requires consistency. Skills policy requires consistency. Regional growth requires consistency. Infrastructure requires consistency. University reform requires consistency. Energy investment requires consistency.

If every policy is rewritten every three years, no serious ecosystem can form.

The test

The test for the next prime minister is not whether they can produce a new slogan for growth.

The test is whether they can answer this question:

What are the good jobs of the future, where will they be, and how will people get into them?

Until that question is answered, Britain will keep producing the same pattern: an overheated London, a handful of successful cities, too many underpowered towns, too many graduates in disappointing jobs, too many small businesses trapped below thresholds, too many high streets fading, and too many people concluding that the economy has no real place for them.

The country does not need more abstract optimism. It needs productive capacity.

It needs regional specialisms. It needs skilled technical routes. It needs serious local employers. It needs high streets that function. It needs universities tied more closely to the industries around them. It needs tax policy that rewards work, mobility and enterprise. It needs a pragmatic relationship with Europe. It needs clean energy policy that builds domestic capability. And it needs a state capable of sticking with a plan for longer than a news cycle.

Britain is not poor because it lacks intelligence or imagination.

It is poor because it has too often failed to turn intelligence and imagination into durable institutions, useful infrastructure, exportable products, skilled jobs and thriving places.

That is the work now.

That last question is the one most designers will jump to, but it is worth starting where Evans starts: with the money.

The numbers are hard to absorb. The big AI labs and hyperscalers are spending extraordinary sums on chips, data centres, energy and infrastructure. This is not the old software model where a small team writes code once and distributes it at near-zero marginal cost. Frontier AI looks much more like a capital-intensive industrial build-out. It needs land, power, cooling, supply chains, GPUs, financing and constant reinvestment.

That creates a strange tension. The companies building large language models want the economics of software, but the cost structure is drifting towards infrastructure. They are spending like telecoms companies or oil majors while hoping to capture value like Google, Microsoft or Apple.

Evans’ implicit question is whether that works.

I think he is right to be sceptical. For most everyday uses, the leading models are starting to feel less differentiated than the companies behind them would like. One might be better at code, another at writing, another at long context, another at multimodal work. But for the average user asking for a summary, a draft email, a spreadsheet formula or a bit of research, the difference is not always obvious enough to create loyalty.

That matters because if the models converge, they risk becoming the mobile carriers of the AI economy: essential, expensive to build, constantly upgraded, but not necessarily where the most attractive margin sits. Nobody doubted that mobile networks mattered. But the real strategic value in mobile went to the operating systems, app stores, devices, payment layers, developer ecosystems and companies that owned the customer relationship.

The same could happen with AI. The model may be necessary without being sufficient.

The model companies clearly understand this. OpenAI does not want to be merely an API provider. Anthropic does not want to be a smarter AWS bill. Google does not want Gemini to sit invisibly beneath someone else’s product. They all want to move up the stack, into the interface, the workflow, the browser, the agent and the place where user intent is captured.

This is where Anthropic is especially interesting. With Claude Code, it is not just selling access to a model. It is building the tool that developers use to do the work. Claude Code can sit inside a codebase, understand the project, make changes, run commands and behave more like a collaborator than a chatbot. That is a different kind of relationship from renting intelligence by the token.

Coding is the obvious beachhead because the output is easier to verify than many other knowledge tasks. The code runs or it does not. The tests pass or they do not. The pull request can be reviewed. The deployment can be rolled back. There is still plenty of judgement involved, but the feedback loops are unusually tight.

The next enterprise question is whether the same pattern extends into design.

That does not mean “make me a pretty homepage” design, although there will be endless amounts of that. It means the broader product work that sits around design: understanding a user need, mapping a workflow, generating interface options, producing prototypes, testing variants, summarising research, writing product specs, and handing the whole thing into code.

If coding becomes cheaper, the bottleneck moves upstream. What should we build? For whom? In what order? With what trade-offs? How should the product behave when the user is confused, impatient, tired, regulated, distracted or trying to do three other things at once?

That is the terrain design and product teams currently occupy. It is also exactly the terrain AI tools are moving into.

Evans is also right that chat is unlikely to be the final interface for most of this. Chat is a useful starting point because it is universal. You can ask anything. You do not need to learn a new interface. It is also a very good replacement for a lot of search behaviour. I already find myself going to ChatGPT or Claude before Google for many questions, especially when I want synthesis rather than a set of links.

In that sense, ChatGPT looks less like a productivity tool and more like a potential Google replacement. It could become the place people land on the internet.

The problem is that search-like behaviour is hard to monetise through subscriptions alone. People did not pay Google to search. They paid with attention, data and a tolerance for advertising. It is not difficult to imagine AI following the same path. Commercial queries get monetised first. Travel, software, shopping, local services, hiring, legal, insurance. The ads may arrive gently, then suddenly feel obvious.

This also explains why browsers have become interesting again. If the assistant becomes the gateway to the internet, the browser is not just a window onto websites. It is the layer that sees what you are reading, what you are trying to do, where you are stuck, what you might need next, and which service should be inserted into the flow. Search was the old command line for the web. The AI browser could become the new one.

But this is also where the strategy gets fragile. If open-source and local models become good enough for a large number of everyday tasks, the general-purpose model layer becomes harder to defend. They do not need to be better than the frontier labs. They only need to be good enough that users and companies stop caring for many common use cases.

That is why the real value may sit elsewhere: in context, proprietary data, distribution, trust, workflow and interface.

This is the part of Evans’ deck I think more founders should sit with. A thin wrapper around a model is not much of a company. It may be a useful feature. It may be a good demo. It may even generate revenue for a while. But unless it owns a workflow, a customer relationship, a data advantage or a behaviour change, it is exposed. The model provider can copy it. The system of record can absorb it. Another startup can rebuild it with the same APIs and a better onboarding flow.

The better opportunity is to understand the work deeply enough that AI becomes part of the work rather than a box next to it.

That means knowing the user’s context. It means having access to the right data. It means understanding the sequence of decisions, exceptions, permissions, anxieties and shortcuts that make up a real workflow. It means designing interfaces that do not simply ask users to prompt better, but help them get the job done.

A lot of the next wave will look familiar. We will get AI-native CRMs, AI-native analytics tools, AI-native customer support systems, AI-native design tools, AI-native research platforms, AI-native legal workflows. Some will be lazy: old software with a magic wand button. Some will genuinely change the unit of work.

That pattern should feel familiar. The mobile revolution did not invent taxis, photography, maps, dating, messaging or hotel rooms. It reorganised existing behaviours around new capabilities: location, cameras, touchscreens, notifications and constant connectivity. AI will do something similar. The interesting products will not simply add generation to an old workflow. They will change what the user thinks the job is.

So far, I am largely with Evans.

Where I start to diverge is around the labour question.

Evans has often taken the historically sensible view that new technology destroys some jobs and creates others. In the middle of the transition, the new jobs are hard to see. This has been true often enough that it deserves to be taken seriously. Mechanisation displaced physical labour. Industrial machines displaced dexterity. Software displaced clerical work. Each time, humans moved up the ladder into planning, judgement, creativity, empathy, coordination and strategy.

The issue is that AI is aimed much closer to the top of the ladder.

Previous waves automated muscle, movement, memory or calculation. AI is starting to automate research, synthesis, writing, planning, coding, interface generation, analysis and decision support. These are not minor tasks for knowledge workers. They are the work itself.

There will be new jobs. Of course there will. We will get AI operations roles, model evaluators, workflow designers, automation auditors, agent supervisors, synthetic data specialists and various new titles that sound faintly ridiculous until suddenly every large company has one. But the real question is not whether new jobs appear. It is whether they appear in anything like the same volume as the jobs being eroded.

I am not sure they will.

This is particularly uncomfortable for designers. I suspect we may have reached peak designer. That does not mean design stops mattering. In some ways, it may matter more. But “design matters more” is not the same as “companies need more designers.”

A lot of design work inside product teams is not the romantic version of the craft. It is assembling flows from a design system, producing variants, documenting states, preparing handoff files, summarising research, creating prototypes, making decks and moving work through a process that exists partly because the tools have historically been too dumb to carry more of the load.

AI weakens that bundle from several directions at once. Product managers will be able to generate credible flows. Engineers will be able to produce acceptable interfaces from existing design systems. Research tools will summarise interviews and customer calls. Experimentation platforms will propose variants and read the results. Design systems will become more generative. The need for a designer on every piece of interface work starts to look less obvious.

The first-order response is that designers move upstream. They own the system, the quality bar, the principles, the interaction patterns and the deeper product choices. I think that is true for the best designers. It is also a smaller job market.

The more uncomfortable version is that some of the system starts to design itself. A product observes behaviour, generates variants, runs tests, analyses outcomes and adapts the interface without waiting for a quarterly planning cycle. The designer does not draw every state. The PM does not imagine every test. The researcher does not manually identify every pattern. The system proposes, tests and adapts.

Humans stay involved, but the loop gets smaller. They set constraints, review exceptions, define brand boundaries, audit for harm and intervene when the system drifts. That might be more interesting work. It may also employ far fewer people.

The same logic applies to product managers. The job contains a lot of synthesis, coordination, prioritisation, documentation, stakeholder management and decision framing. AI can already help with much of that. The best PMs may become much more effective. The average PM may become less necessary.

Engineering may be protected for a while by backlog. Most companies have more software they want built than they have people to build it. AI coding tools will initially look like a pressure valve. Old tickets get cleared. Internal tools finally get made. Integrations that were never worth the effort suddenly become possible.

But once the backlog has been eaten, the labour question returns. If one engineer with AI can do the work of three, do companies build three times as much software forever? Some will. Many will not. At some point, the CFO notices.

This is where the “technology always creates more jobs” argument feels a little too comforting. It may still be true in the long run. But for many existing roles, the practical effect may be a shrinking of the team rather than a flowering of new adjacent careers.

AI may not replace whole professions in one dramatic motion. It is more likely to erode the task bundles that justify headcount. A designer is not replaced by “an AI designer.” A designer is squeezed by a design system, an AI prototyping tool, a PM who can generate flows, an engineer who can produce decent UI, a research synthesis tool, an experimentation platform and a leadership team willing to accept good enough for most surfaces.

That is a less cinematic story than mass replacement, but probably a more realistic one.

This is why Evans’ best question is also the most useful one for product and design leaders: what becomes cheap?

Cheap code changes the economics of software. Cheap content changes marketing. Cheap analysis changes management. Cheap interface generation changes product teams. Cheap research synthesis changes strategy. Cheap experimentation changes decision-making.

The follow-up question is more uncomfortable: which expensive organisational habits survive once those things are cheap?

Large product teams are partly a response to scarcity. Scarce engineering time. Scarce design production. Scarce research capacity. Scarce analytical attention. Scarce coordination bandwidth. AI does not remove all of those constraints, but it attacks enough of them that the old team shape starts to look exposed.

The optimistic version is that teams become smaller and better. Less time spent producing artefacts for each other. More time spent understanding customers, making difficult product choices and building things that previously sat below the economic threshold. The pessimistic version is that companies use AI to thin teams, flood the world with adequate software and mistake automated optimisation for product judgement.

Both will happen.

For designers, the lesson from Evans’ deck is not simply that AI is a platform shift. That much is obvious. The more useful lesson is that platform shifts rearrange where value sits. In mobile, value moved from carriers to operating systems, apps and services. In AI, it may move from models to workflows, context, data and interfaces. Inside companies, it may move from production to judgement.

That should make designers both more confident and more nervous.

More confident because design, in the broadest sense, is exactly what makes AI useful: understanding context, shaping workflows, making systems legible, deciding what good looks like, and preventing technically impressive products from becoming unusable mush.

More nervous because many of the artefacts designers have historically produced are about to become cheap.

If the model layer becomes infrastructure, the product layer matters more. If chat is a transitional interface, workflow design matters more. If generation becomes abundant, taste and judgement matter more. But none of that guarantees the current shape of the design profession survives intact.

The safest place to stand is not inside the artefact. It is closer to the decision.

The automotive industry has a useful way of talking about autonomy, from basic driver assistance through to fully self-driving cars. Software is likely to develop a similar ladder. Not around who has their hands on the wheel, but around who decides how the product changes. Here's my attempt to shape what that ladder might look like, starting with the way we work today.

Level 0: Manually Authored Software

This is the way most software has been designed up to this point.

A team does research. They look at customer interviews, usability studies, support tickets, sales calls, analytics dashboards, funnel reports, session replays, app store reviews and whatever other signals they can gather. Product managers, designers, researchers, analysts and business stakeholders interpret that evidence, turn it into insights, then propose improvements.

Those ideas become features, experiments, backlog items, design briefs, roadmap themes or quarterly priorities. The product team designs the solution, engineers build it, QA tests it, and the organisation releases it. If the team is reasonably disciplined, someone later checks whether the change had the desired effect.

Level 0 software may be heavily instrumented, but the loop is still stitched together by people. Analytics tools do not redesign the checkout flow. Support tools do not patch the confusing settings page. Session replay does not generate a better onboarding journey. These tools produce evidence, but humans decide what matters and push the change into production.

Core loop:
Research and data → human interpretation → human concept → human design → human build → human testing → human release → human measurement

Example:
A subscription product has separate checkout flows for the UK, Germany and the US. Each flow is researched, designed, localised, built, tested and optimised manually. Analytics may show that one market is underperforming, but the team still has to understand why, decide what to change, and ship the improvement.

Level 1: AI-Assisted Software Creation

This is the level most teams are currently moving towards. Humans still decide what to build, but AI helps them think, design and ship faster.

This is the current wave of AI coding tools, design agents, prompt-to-app builders, synthetic users, automated QA, AI-generated copy, AI-generated prototypes and Figma-to-code workflows. Product managers and designers use LLMs to explore ideas, simulate reactions, critique flows, draft experiments and imagine product improvements. Engineers use tools like Cursor, Copilot and coding agents to move faster through the backlog.

This changes the pace of product work. More ideas can be explored. More prototypes can be generated. More tickets can be completed. More experiments can be shipped. In some teams, the bottleneck moves from “can we build this?” to “should we build this, and did it work?”

A lot of product teams, design teams and AI-native studios currently treat this as the destination. In their view, the future of software is smaller teams using AI agents to design, code and ship at a speed that previously required a much larger organisation.

That is a meaningful shift, but it is still a human-orchestrated model. The product team is using AI to make software faster. The software itself is not yet observing, diagnosing, repairing or adapting its own behaviour.

The loop is still not closed. In most organisations, the context needed to make good product decisions is scattered across analytics dashboards, support tools, CRM notes, research repositories, experiment logs, sales calls, roadmap docs and half-forgotten Slack threads. Much of the important context still sits in people’s heads: why a decision was made, which metric can be trusted, which customer segment matters most, which old experiment failed, and which apparent opportunity is actually a trap.

So even when humans are doing less hands-on coding, they are still the glue. They carry the context, interpret the evidence, make the trade-offs, decide what gets shipped and check whether the result was any good.

Core loop:
Human intent → AI-assisted exploration → AI-assisted production → human review → human deployment → human measurement

Example:
A product manager uses an LLM to brainstorm improvements to a subscription checkout, then the team uses AI coding tools to build several variants more quickly. Humans still decide what to test, connect the work to the actual analytics, judge whether it succeeded and decide what happens next.

Level 2: Self-Diagnosing Software

This is the point where a product starts to develop a sense of proprioception.

In humans, proprioception is the sense of where your body is, how it is moving, and when something feels wrong. Level 2 software begins to develop something similar. It starts to understand how it works, where it is under strain, when normal behaviour has changed, and where somebody may need to intervene.

This is the first point where adaptive software feels meaningfully different from today’s product process. The system no longer waits for a product manager, analyst, researcher or growth lead to open a dashboard and go looking for problems. Agents run in the background, constantly watching what is happening across the product.

They monitor usage patterns, funnel performance, error logs, support tickets, session replays, release history, experiment results, performance data and customer behaviour. More importantly, they connect those signals together.

At this level, the system may also start flagging opportunities. It might notice that expert users are slowed down by explanatory copy, that one onboarding step is consistently skipped, that search is being used as a workaround for poor navigation, or that users in one market abandon after seeing a payment method they do not trust.

It may even suggest possible fixes or experiments. But the boundary remains clear: the software diagnoses and recommends. Humans still decide whether the diagnosis is credible, whether the opportunity matters, and whether to act on it.

Core loop:
Product observes → product connects signals → product diagnoses → product recommends → human decides

Example:
A subscription product has agents monitoring checkout behaviour across markets. The system notices that German mobile users are abandoning after postcode entry, connects this to a recent release, checks related support tickets and session replays, then flags the issue with a likely cause and possible tests.

Level 3: Self-Healing Software

At level 3 the product can detect and fix narrow failures automatically.

This is where the system stops simply flagging problems and starts making small, safe corrections. It is still not redesigning the product. It is not deciding strategy, inventing new features or rethinking the user journey. It is keeping the existing product healthy.

To do that, the system needs some understanding of developer intent. It can infer that intent from the codebase, unit tests, API documentation, design system rules, FAQs, acceptance criteria, release notes, monitoring thresholds and the behaviour of previous working versions. It starts to understand not just what the software is doing, but what it appears to be meant to do.

A Level 3 system constantly runs checks in the background. It can detect failing tests, broken components, visual regressions, slow services, payment-provider issues, accessibility errors, dependency problems, dead links, API failures, layout glitches, degraded performance, emergent bugs and small security vulnerabilities.

At the simplest level, it may roll back to the last known working version. That alone would remove a lot of stress from engineering teams. But increasingly, the system may be able to fix the issue directly: patching a dependency, correcting a validation rule, restoring a broken API contract, repairing a visual regression, or opening a tightly scoped pull request with the likely cause already explained.

The analogy is less “fully self-driving car” and more modern autopilot in an aircraft: hundreds of small adjustments in response to pressure, wind, turbulence and changing conditions so the flight stays smooth. Or, in car terms, automatic emergency braking. The system is not doing all the driving, but it is watching continuously and reacting faster than a human might to critical issues.

Self-healing software is defensive. It maintains the intended experience rather than inventing a better one. It keeps the system inside safe operating conditions and frees engineers from some of the low-level bug-fixing and technical-debt work that currently eats into product development.

Core loop:
Product observes → product tests → product detects failure → product infers intended behaviour → product repairs or rolls back → human is notified

Example:
A new checkout release creates a spike in failed payments for users in France. The system detects the issue, compares it with tests and previous behaviour, identifies a broken payment-provider integration, switches affected users back to the previous component, opens a pull request with a proposed fix, and notifies the engineering team.

Level 4: Self-Optimising Software

At this level the product starts to generate, run and evaluate improvements to known flows.

This is where adaptive software moves beyond maintenance and starts doing some of the work we currently associate with product, growth and experimentation teams. The system is no longer just keeping the product healthy. It is trying to make the product perform better.

In the early stages, this will probably be supervised. The software might notice a problem, generate a set of possible experiments, and ask a human to authorise the test:

“German mobile users are abandoning at postcode entry. I’ve generated three alternative flows: one with later address validation, one with clearer invoice-payment copy, and one that moves payment selection earlier. Shall I run these as a 10% experiment?”

Once teams trust the system, some of this work can move into autonomous mode. The product can run multivariate tests in the background, allocate traffic, monitor performance, stop harmful variants, promote winners and keep iterating without asking for permission every time.

At this level, the product still has one main version of the software. It is not creating a different interface for every user or designing around dozens of behavioural clusters. It is optimising a known flow toward a single local maximum: the best checkout, the best onboarding sequence, the best upgrade path, the best cancellation save, the best pricing page for an "average user".

The difference is that it can optimise continuously, across more dimensions than a human team could realistically hold in their heads. A mature Level 4 system would understand business goals such as acquisition, activation, conversion, revenue, retention, support cost, basket size, lifetime value, churn risk, margin and customer satisfaction. It would know that improving conversion at the expense of refunds, support load or retention may not be an improvement at all.

This is where governance starts to matter. A self-optimising system needs more than a metric to chase. It needs constraints, trade-offs and a working definition of what “better” is allowed to mean.

Core loop:
Product diagnoses → product generates variants → product runs experiments → product evaluates outcomes → product ships winners → product keeps optimising

Example:
The system identifies friction in a subscription checkout, generates several variants, runs a multivariate test, notices that one version increases conversion but also increases refunds and support contact, rejects that variant, then promotes a slightly lower-converting version that produces better long-term retention and lower support cost.

Level 5: Multi-Optimum Software

Products at this level start to create multiple versions of itself for different groups of users, needs and contexts.

Once a system can diagnose problems, generate variants, run tests and optimise known flows, the next step is to stop assuming there is one best version of the product. Instead of optimising a single checkout, onboarding journey or product page toward one local maximum, the software can start creating multiple local maximums for different types of user.

At first, this might look familiar. The system could automatically design different checkout flows for different geographies: one for Germany, one for the US, one for Japan, one for Brazil. Each might reflect different payment norms, address formats, tax rules, trust signals, delivery expectations and regulatory constraints.

But geography is only the obvious example. An ecommerce product might discover that older customers behave differently from younger customers around payment, reassurance, font size, product detail, returns information or customer support. It might learn that some users prefer card payment, while others expect wallets, bank transfer, invoice payment or instalments. It might find that different groups respond better to lifestyle imagery, technical specifications, comparison tools, video, reviews, sizing guidance, augmented reality previews or social proof.

This is where things start to get weird. There is no longer a single source of truth in the traditional product-design sense. You cannot print out the user flows, stick them on the wall, and say, “This is how the product works.” The product may have 20, 50 or 200 active versions, each optimised for a different cluster of users, needs and contexts.

At that point, teams may stop fully understanding the product in the old way. They will understand the system, the goals, the constraints, the component library and the rules of adaptation. But they may not know every expression of the product that users are seeing. The product becomes less like a designed artefact and more like a managed ecosystem.

This is still bucketed personalisation. Users are being assigned to clusters. But the clusters are increasingly discovered by the system rather than invented in a workshop. The product is no longer asking, “What is the best version of this flow?” It is asking, “Best for whom?”

That creates a serious governance challenge. Some categories, such as age, gender, geography or inferred intent, can quickly become sensitive or reductive if used crudely. A mature Level 5 system cannot simply optimise whatever appears to work. It needs rules about fairness, accessibility, manipulation, exclusion, brand consistency and what kinds of personalisation are acceptable.

Core loop:
Product identifies clusters → product generates tailored journeys → product routes users → product evaluates outcomes by cluster → product refines the experience set

Example:
A shopping site discovers that different groups need different checkout and product-page experiences. It automatically creates one checkout for invoice-heavy B2B buyers, another for younger mobile-wallet users, another for older customers who need more reassurance, and another for users in markets where delivery trust is the main barrier.

Level 6: Individually Adaptive Software

Products at this level have the ability to create a different a difference experience for every user, interaction or session.

Once a system can generate dozens or hundreds of versions of a product for different groups, the final boundary starts to look arbitrary. If the product can adapt to geography, payment preference, confidence level, device context, previous behaviour, buying intent, accessibility need and stage in the journey, why stop at buckets?

At Level 6, the product no longer routes users into a pre-existing segment and serves the best available version for that group. It assembles the interface around this user, in this moment, for this task.

The same person might get a different version of the product on Monday morning than they do on Friday evening. They might get a different interface when browsing casually on a phone, comparing options on a laptop, returning after abandoning a basket, buying under time pressure, using assistive technology, or contacting support after something has gone wrong.

At this level, every interaction becomes its own design brief.

The system is not choosing between Checkout A, Checkout B or Checkout C. It is dynamically assembling the journey from components, content, interaction patterns, business rules, accessibility requirements, brand principles, regulatory constraints and live behavioural signals. Sequence, density, copy, reassurance, navigation, visual hierarchy, defaults, support prompts and payment options can all shift according to the user’s needs and context.

In Level 5, the system creates local maximums for groups. In Level 6, it tries to create a local maximum for every individual interaction.

That could be extraordinarily useful. A novice user gets more explanation. An expert gets shortcuts. A nervous buyer gets reassurance. A returning user skips steps. A user with accessibility needs gets a structurally different interface. A user in a regulated market gets additional compliance content without making the product heavier for everyone else.

It could also become manipulative very quickly. A product that can adapt to someone’s needs can also adapt to their weaknesses. So Level 6 is not just a technical endpoint. It is a governance problem. The question is no longer only, “Can the product adapt?” It is, “What is it allowed to adapt around, and in whose interest?”

Core loop:
Product understands the individual context → product assembles the experience → product evaluates the outcome → product updates its model → product adapts again

Example:
Two people arrive at the same subscription checkout and receive meaningfully different experiences: different sequence, density, explanation, payment emphasis, reassurance, support prompts and interaction patterns. Later, one of those same users returns on a different device, with a different intent, and receives a different version again.

The underlying shift

The future of software is not simply that teams will use AI to build products faster. That is Level 1.

The larger shift is that products themselves will begin to observe, diagnose, repair, optimise and eventually reshape their own interfaces.

Most software today is designed around a small number of assumed use cases. This requires trade-offs and those trade-offs require human input. Adaptive software moves from one local maximum, to many local maximums, to a local maximum for every individual user interaction. And this all happens live, in the background, and without needing additional resource.

When this happens, product development stops being a human endeavour and becomes a "lights-off" activity, much like factories or warehouses that have such a high level of automation, they can run mostly in the dark, because humans rarely if ever need to visit the factory or warehouse floor.

Not every product team will operate like this, just like not every car will be self driving in 10 years time. But this is defnitly the direction of travel and I'm already starting to see products hitting levels 3, 4, 5 and crude versions of 6.

No new Labour leader is going to walk into Downing Street and visibly transform the country in the time available. Some of the government’s problems are self-inflicted. Some are failures of leadership. Some are the result of a brutal economic inheritance. Some are simply the reality of government: meaningful change takes time, often happens in unglamorous policy areas, and rarely cuts through quickly enough to shift the public mood.

The danger is that Labour replaces Starmer too early, a new leader inherits the same constraints, spends the next couple of years being blamed for not fixing them, and then arrives at the next election looking like the second failed Labour prime minister rather than the fresh start the party needs. Labour would not have one damaged leader. It would have two.

A better, colder strategy would be for Starmer to stay in place for now. Let him absorb the blame. Let the media, the opposition and the public focus their anger on him personally. Let the story become: this was a Keir Starmer problem, rather than proof that Labour itself has nothing left to offer.

Meanwhile, Labour should start elevating its next generation. Put them in visible jobs. Let them front the areas where the government is making progress. Let them own the policies that work. Let them build relationships with the public before they are formally asking for the keys. Then, closer to the election, Labour can make the break. Starmer goes. A new leader comes in with enough time to define themselves, but not so much time that they become fully contaminated by the old government’s failures.

At that point, the pitch has to feel like a different Labour project: more optimistic, more patriotic, more emotionally alive. It cannot be a reheated New Labour tribute act or another managerial promise to “deliver”. It needs to be a serious modern story about national renewal: Britain as a creative, open, inventive middle power with strong institutions, world-class universities, cultural reach, industrial potential and a future that does not have to look like managed decline.

That requires Labour to be much clearer about what it is for. And it has to begin with the emotional claim Reform has been allowed to own: Britain looks after other people better than it looks after you.

That is the story Reform tells, in one form or another. Sometimes explicitly, sometimes by implication. It sits underneath the anti-immigration rhetoric, the anger about hotels, the suspicion of London, the resentment towards universities, the attacks on welfare, the contempt for liberal institutions, and the sense that every other group has a claim on the country except the people who have lived here all along.

I don’t think that story is true. But it is effective because it attaches itself to things people can see. A lot of people who have moved from Labour or the Conservatives to Reform are not bad people. Many are not especially ideological. They are angry, disappointed and tired of being told things are getting better when the place they live tells them otherwise every time they walk down the high street.

They see boarded-up shops. They see betting shops, vape shops, charity shops and empty units where there used to be banks, butchers, cafés and useful local businesses. They see coastal towns that once had civic pride now looking as though they have been left to rot. They see former industrial areas where every promise of regeneration seems to arrive as a press release and disappear as a planning dispute. Into that gap, Reform offers its simple explanation: you have been forgotten, others have been put ahead of you, and Britain no longer belongs to you.

Labour cannot rebut that with statistics. It cannot scold people out of feeling abandoned. It cannot tell them that their lived experience is wrong because some national indicator has moved in the right direction. It has to answer with visible change.

This is where the next Labour project needs to begin: with the places people feel have been neglected for too long. Not with another vague regeneration fund, another logo, or another minister in a hard hat pointing at a laminated board showing a future transport hub. With work people can actually point to.

That means repaired high streets, cleaner public spaces, better buses, reopened community buildings, support for independent businesses, grants to bring empty units back into use, clinics people can access, markets that feel alive, maintained parks, public toilets, libraries, street lighting and youth centres. These are the things that sound mundane in a manifesto but change how people feel about where they live.

Some of this needs to happen quickly. A lick of paint does not solve structural inequality, but visible neglect creates political despair. When people live somewhere that looks abandoned, they conclude that they have been abandoned. Labour needs to make the first signs of repair impossible to miss.

The irony, of course, is that many of the things Labour now needs to make visible are exactly the sorts of things European regional funding used to help support before Brexit: public realm improvements, local infrastructure, business support, skills programmes, community facilities, environmental work and the quiet civic maintenance that made neglected places feel noticed. The signs with the EU flag on them may have been easy to mock, but they often marked something real: a rebuilt square, a training programme, a business centre, a restored public building, a local project that would otherwise have struggled to happen.

That is not an argument for re-running 2016. It is a reminder of what was lost when Britain chose to stand apart from European structures that had, imperfectly but materially, invested in some of its most overlooked places. Labour should be careful not to turn this into a “you voted against your own interests” lecture. That would be fatal. It can say something more useful: the old model has gone, and the promised replacement has not been good enough. If we are serious about national renewal, Britain now has to rebuild that capacity ourselves.

There is also a tax story here. At the moment, too much of the burden falls on physical businesses trying to keep local places alive. A small shop in a struggling town pays rent, business rates, energy bills, staff costs and local overheads, while large digital businesses can extract value from those same communities without contributing in anything like the same visible way.

The UK already has a Digital Services Tax, charged at 2% on the revenues of large search engines, social media services and online marketplaces that derive value from UK users. Labour could turn this into a much clearer political argument: if we want thriving town centres, we cannot keep taxing the physical high street as if the internet never happened. This is not about punishing technology companies for being successful. It is about rebalancing the system so local businesses are not carrying a disproportionate share of the cost of maintaining the places we all depend on.

Reform feeds off decline. Labour has to become the party that repairs it, without pretending every town can be transformed overnight or promising some twee market-square fantasy. The promise has to be credible and visible: these places are not being written off.

The NHS has to sit inside that same argument. Most British voters, including many on the right, still have a deep emotional attachment to it. Labour should stop talking about the NHS only in operational terms: waiting lists, reform plans, productivity targets. Those things matter, but they do not touch the deeper truth, which is that the NHS is part of who we are.

That is where Labour can draw one of its strongest contrasts with Reform. Nobody thinks the NHS is perfect. Everybody knows it needs repair. But the choice is between rebuilding a public institution most people love and allowing it to be hollowed out, outsourced and pushed towards an American-style model that very few British people actually want.

The same is true of welfare. Labour does not need to defend every part of the current system uncritically, but it does need to separate reform from cruelty. A lot of right-leaning voters would be surprised by where Reform’s instincts lead when you move past the slogans. The public may be sceptical of welfare abuse, but that is not the same as wanting people with serious disabilities, long-term illness or genuine need treated as a burden. Labour should be much more confident here. The system should be fair, and people who need support should not be thrown to the wolves.

Only once Labour has taken people’s immediate sense of neglect seriously can it talk honestly about Brexit. Because Labour still does not know how to do that.

The worst version is easy to imagine: “You were lied to. You got it wrong. Now admit it.” That would be politically useless. Millions of people voted Leave for reasons that were sincere, emotional and often rooted in real frustration. They wanted more control. They wanted the country to feel less remote. They wanted politics to answer to them rather than to institutions they did not trust. Labour will get nowhere by making those voters feel stupid.

But it also cannot keep pretending Brexit is working. The right way to talk about Brexit is economic rather than moral. Britain made a big economic bet on the shape of the world, and that bet has not paid off.

The hope was that leaving the EU would make Britain a more agile global trading nation. Less tied to European rules. More open to the wider world. Better able to strike deals with America, Asia and emerging markets. But the world has moved in the opposite direction. America has become more protectionist. China has become more dominant. Global trade has become more fragmented. Supply chains have become more political. In that world, being outside your nearest major trading bloc has made Britain more exposed.

The Office for Budget Responsibility assumes that Brexit will reduce long-run UK productivity by around 4% compared with remaining in the EU, and that both imports and exports will be around 15% lower in the long run than they otherwise would have been. Labour should not lead with GDP charts, though. Most people do not experience GDP. They experience prices, jobs, paperwork, delays, lost orders, weaker public finances and the general sense that the country has less room to move.

The language needs to be simpler. We tried a big national experiment. We were told it would make us more prosperous. It hasn’t. Britain is poorer than it needed to be, and some parts of the economy have paid a much higher price than others.

For people in London working in finance, tech or professional services, Brexit has often felt manageable. Annoying, perhaps. Occasionally expensive. But not existential. For manufacturers, exporters, food producers, farmers and smaller businesses trading with Europe, it has been much harder. If you make specialist components in the Midlands, sell cheese from Somerset, build machinery in Yorkshire, export shellfish from Scotland, or run a small business that depends on European customers, Brexit did not feel like sovereignty. It felt like forms, delays, costs and lost trade with the people closest to us.

A future Labour leader should be able to say that we are not going back to the arguments of 2016, and we are not asking people to relitigate how they voted. But we do have to be honest about what has happened since. Britain is poorer outside the European trading system than we were told it would be. The world has changed. Our economy needs a closer, more modern relationship with Europe.

That phrase matters: a closer, more modern relationship with Europe. The point is not to pretend the last decade did not happen, or to suggest Britain can simply walk back in and demand all the benefits of the old arrangement without the obligations. Any new deal will probably be worse than the one Britain used to have. We gave up a uniquely strong position. But a worse deal than the one we had is still better than refusing to improve the one we have now.

This is also where Labour can make Reform look smaller. Reform’s answer is to double down on isolation. Labour’s answer should be to repair the damage in the national interest. British businesses need to trade. British workers need good jobs. British manufacturers need supply chains that work. British farmers and food producers need easier access to their nearest customers. British families need a country that is richer, more stable and less alone.

Brexit was sold as a way to make Britain stronger. Making Britain stronger now means rebuilding the relationships Brexit weakened.

This is about more than trade. It is also about the kind of economy Britain wants to build. For too long, national growth has been imagined through a narrow set of places and industries: London, Manchester, finance, professional services, tech, property, hospitality, consumption. Those sectors matter, but they cannot carry the whole country.

If Labour is serious about renewing towns outside the great service-economy centres, it needs an industrial story that reaches beyond coffee shops, logistics sheds and call centres. That means advanced manufacturing, clean energy, engineering, climate technology, creative technology, medical devices, biotech, robotics, defence, public transport, housing retrofit and AI applied to real-world industries rather than just software companies selling tools to other software companies.

There is a skills story here too. Town renewal cannot simply be about beautification. It has to connect to education, apprenticeships, technical colleges, universities, local employers and new industries. If you want people to believe their town has a future, there need to be decent jobs within reach. Not jobs that technically exist somewhere in the region, but jobs people can imagine their children doing without having to leave and never come back.

That is where a better relationship with Europe matters again. Closer ties with Europe would make it easier for manufacturers to trade, for universities to collaborate, for researchers to work across borders, for creative companies to reach audiences, for engineering firms to plug into supply chains and for British towns to participate in an economy bigger than the domestic market alone. Europe is not a distraction from national renewal. It is one of the conditions that makes national renewal easier.

Once Labour has made the practical argument, it can start to make the bigger one: Britain needs a new account of its place in the world. This is where Labour has to reclaim a language of national pride.

One of the mistakes progressives make is assuming that visible patriotism is automatically reactionary. It isn’t. A lot of people putting up flags across the country are not doing it because they hate anyone. They are doing it because they want Britain to mean something. They want to love the place they are from. They want a story about the country that feels bigger than decline.

The problem is that the story they are being offered by Reform and parts of the Conservative Party is cramped and backward-looking: Churchill, the war, the Blitz, empire, standing alone, a version of Britain permanently lit by the glow of past achievement. There is nothing wrong with honouring courage, sacrifice or national resilience. But when patriotism gets trapped there, it becomes less about what Britain might become and more about defending a selective memory of what Britain once was.

That is how you end up with a culture war around the National Trust. The problem is not that the National Trust hates Britain. It is that it takes the country’s history seriously enough to look at all of it: the country houses, the gardens, the philanthropy, the art, the exploitation, the slave trade, the wealth and the violence that often sat behind the polished stone. A mature country should be able to handle that.

Labour needs to offer a more adult patriotism, one that does not sneer at national pride or build it on denial. There is another British story available: public service, libraries, parks, universities, the NHS, the BBC, local government, mutual aid, trade unions, scientific discovery, engineering, design, theatre, music, literature and law. A story about a country that has often been at its best when it has built institutions bigger than the individual, and then invited the world to learn from them.

Britain helped shape the industrial revolution. It built railways, factories, universities, civic museums, public libraries and municipal infrastructure. It also produced many of the movements that fought to humanise the consequences of that industrial power: labour rights, public health, social housing, universal education, the welfare state. That is a far richer source of national pride than nostalgia for empire, and it points forwards. If Britain once helped shape the industrial revolution, the question now is how it shapes the next one.

Britain is no longer an imperial power. It is not a superpower. It cannot pretend the world works as it did in the 19th century, or even the late 20th. But it remains a country with extraordinary cultural reach. British music, film, television, theatre, publishing, comedy, fashion, games, design and journalism still travel around the world. Some of the most admired shows, songs, books, performances and creative companies have been shaped by British talent. Culture is one of the main ways Britain makes itself felt in the world.

It is also one of the places where British values are carried most effectively. Not through ministerial speeches, but through stories, humour, language, drama, criticism, invention and taste. That is why institutions like the BBC matter so much. The BBC is not perfect. No national institution is. But the constant political habit of doing it down has become self-defeating. It is one of Britain’s most recognisable cultural institutions, one of its most trusted global brands, and one of the great nurseries of British creative talent. Writers, producers, actors, directors, presenters, journalists, composers, editors and technicians have passed through it, learned their craft, and gone on to shape culture far beyond these islands.

Defending the BBC should not be treated as a narrow culture-war position. It should be part of an industrial strategy for British creativity. The same is true of theatres, libraries, universities, museums, art schools, music venues, public service broadcasting and local cultural spaces. These are part of the machinery through which Britain produces the thing it is still unusually good at producing: culture that travels.

Our universities matter for the same reason. They are among Britain’s great national assets, producing research, talent and economic value while also projecting British values into the world. People come here to study law, medicine, politics, engineering, design, economics, literature and public policy. They encounter a liberal democracy, imperfect but real. They build relationships. They absorb habits of argument, criticism, institutional life and civic freedom. Many take those experiences back into government, business, civil society and culture around the world.

That should be a source of pride. Instead, universities have been dragged into the immigration argument as if international students are simply a number to be reduced. International students are not just a funding mechanism, though in practice they do help keep many universities afloat. They are also one of the ways Britain remains connected, influential and admired.

Immigration has to be managed. Communities need infrastructure. Housing, schools, transport and healthcare need to keep up. But Britain’s openness to students, researchers, artists, entrepreneurs and skilled workers has been one of the reasons the country has continued to matter. A confident country does not shut the door on people who want to learn here, build here, create here and carry some version of Britain back into the world.

This is the kind of national pride Labour should be reaching for: Britain as a modern European country, open, creative, democratic, institutionally serious, internationally connected, and honest about both its strengths and its limits. The next version of Britain will not be built by pretending the past can be restored. It will be built by asking where the country is genuinely strong now, where it can become stronger, and what role it wants to play in a more unstable world.

There is another part of this patriotic argument Labour should be much more willing to make. Defending Britain means defending British democracy.

For years, Britain has been too relaxed about Russian money, Russian influence and the way hostile states seek to shape public life without having to invade or occupy anything. This is not about seeing Kremlin plots behind every bad opinion. Labour needs to be careful here. Overclaiming would be easy and politically stupid. Underplaying the problem would be just as foolish.

The Intelligence and Security Committee’s Russia report warned that the UK had become one of Russia’s top intelligence targets, and that Russian influence in the UK had become “the new normal”. The House of Commons Library summarised the report as saying that Russian oligarchs had used business interests, donations to charities and donations to political parties to influence UK affairs.

Then there is Nathan Gill, the former Reform UK leader in Wales and former MEP, who was jailed for taking bribes to make pro-Russian statements while he was in the European Parliament. He pleaded guilty to eight counts of bribery relating to activity between December 2018 and July 2019. Labour should not pretend this proves every Reform voter has sympathy for Putin. That would be absurd and offensive. But it does raise a fair question about the kind of politics Reform has helped normalise.

Why are the loudest self-declared patriots so often relaxed about foreign influence when it comes dressed in the right sort of grievance? Why does their suspicion of elites seem to disappear when the money, media ecosystem or political style comes from people hostile to Britain’s democratic interests? Why is the flag treated as sacred, while the institutions that protect the country are treated as disposable?

Real patriotism is not just waving a flag. It is defending the courts, the civil service, Parliament, independent journalism, public service broadcasting, universities, the security services, the rule of law and the integrity of elections. It is taking hostile influence seriously, whether it comes through money, media, lobbying, disinformation or useful idiots repeating talking points they barely understand.

Labour should say plainly that Britain’s democracy is part of what makes the country worth defending. If hostile states want to weaken Britain, they do not need to defeat it militarily. They only need to make people distrust every institution that holds the country together. That is why attacks on the BBC, the courts, universities, the civil service and Parliament should not be treated as separate culture-war skirmishes. Together, they form a pattern: a slow erosion of the things that make Britain governable, credible and free.

A serious patriotism would defend those things, while still accepting they need reform. A country which casually destroys its own foundations is not becoming more sovereign. It is becoming easier to manipulate.

The same patriotic frame can be used against Reform’s relationship with America’s hard right. Labour should be much more willing to draw a line between Britain and Trump’s America, and it should do this in patriotic rather than smug liberal terms.

We do not want American-style healthcare. We do not want American-style politics. We do not want public life governed by billionaires, conspiracy theories and permanent grievance. We do not want Britain turned into a tribute act for the American alt-right. That is the trap for Nigel Farage. Make him own it.

Reform wants to present itself as the party of common-sense patriotism. Labour should challenge that directly. Is it patriotic to weaken the NHS? Is it patriotic to leave British manufacturers with more barriers to their nearest markets? Is it patriotic to import the worst instincts of American politics? Is it patriotic to exploit the decline of struggling towns while offering little more than anger in return? Is it patriotic to weaken the institutions that make Britain harder for hostile states to manipulate?

Labour should not tell Reform voters they have been conned. That will only make people defensive. It should show them what Reform really means in practice: your NHS less safe, your town no better off, your local businesses still struggling, your country more isolated, your politics nastier, your public services weaker, your future sold as sovereignty while power moves further away from you.

Then Labour needs to offer something better than managerial competence, technocratic caution or another promise that the adults are back in the room. It needs a project.

Repair the towns. Rebuild the NHS. Reconnect with Europe. Defend the institutions that make Britain worth belonging to: the BBC, the NHS, the National Trust, our courts, libraries, universities, museums and parks. Back manufacturing. Fund the creative industries. Support small businesses. Treat neglected places as national assets rather than electoral problems. Make Britain feel confident without making it cruel.

That is the emotional territory Labour should want to occupy. The party does not need to become more left-wing in some abstract factional sense, and it does not need to chase Reform into performative cruelty. It needs to become more visible, more concrete and more willing to talk about national pride in a way that does not sound embarrassed by the country.

This is why replacing Starmer now may be the wrong move. The next Labour leader needs to arrive carrying renewal, not inheritance. They need to feel like the answer to the failure, rather than another stage of it.

Starmer may be competent at parts of government. He may be better suited to international diplomacy than some of his domestic critics allow. He may be capable of incremental policy work that matters more than the headlines suggest. But he does not create confidence. He does not make the country feel larger, braver or more imaginative. He does not give people the sense that Britain is entering a new chapter.

Labour needs someone who can speak to Reform-curious voters without contempt, to Green-curious voters without defensiveness, and to the wider country without sounding embarrassed by ambition. It needs a leader who can say that Britain has been badly run without implying that Britain is finished. Someone who can talk about flags without sounding frightened of them, immigration without cruelty, Europe without reopening every wound from 2016, and the NHS, the BBC, universities, manufacturing, science and culture as parts of the same national story.

It needs inspiration. It needs hope. It needs a leader capable of making people feel that the country can become more than the sum of its current frustrations. That person has to come from the next generation of Labour leadership.

So let Starmer carry the failure for a while longer. Let him absorb the anger. Let him become the symbol of the caution, drift and disappointment of the first phase of this government. But do not waste the time. Use it to build the next story. Use it to elevate the next generation. Use it to make visible repairs. Use it to prepare a more honest argument about Brexit, Reform, the NHS, neglected towns, British culture, British institutions, hostile influence and Britain’s place in the world.

Then make the break.

The next election will not be won by pretending the last few years have gone well. They haven’t. It will be won by separating Labour’s future from Starmer’s failure, while giving voters something more hopeful than Reform’s politics of grievance.

Britain does not need to be told it is broken. People know what they can see. The task is to show them it can be repaired, and then give them a reason to believe it can become something more.

It’s a neat argument, which is usually a warning sign. Europe clearly has problems. It has fragmented markets, less late-stage capital, weaker stock option culture, slower public procurement, fewer repeat scale-up operators, and a much smaller pool of people who have been through the journey of building genuinely huge technology companies. So yes, it is harder to build a Google-scale company in Europe than in the US. But “Europe is over-regulated” often feels like one of those phrases people reach for when they want to sound serious without doing the work.

Ask a simple follow-up and the argument often starts to wobble: which regulation, specifically? Not vibes. Not “red tape”. Not a general sense that Brussels is annoying. Which actual rule is stopping European founders from building great companies? The answers are often surprisingly vague. Some people mention labour laws. Some mention GDPR. Some point to environmental rules, food standards, packaging rules, employment protections, or the fact that it’s harder to fire people in Europe than it is in parts of the US. But then you look at the claim being made and it starts to feel odd. Is America’s advantage really that workers there get fewer holidays? Is Europe falling behind because we have stronger food safety standards? Is GDPR annoying? Absolutely. Does it explain Silicon Valley? I’m not convinced.

Markets are made of rules

One of the things I find strange about the deregulatory argument is that it seems to imagine some kind of natural state of capitalism: companies simply building, selling, hiring and growing, until governments came along and started throwing sand into the gears. But markets don’t exist before regulation. In many ways, they are regulation.

Contracts, courts, tax systems, corporate law, employment law, property rights, accounting rules, banking systems, liability, insurance, bankruptcy, competition law and product standards are not decorative extras. They are the scaffolding that allows markets to function in the first place. Every advanced economy is regulated. The interesting bit is who the rules serve.

This is where the simplistic critique of Europe starts to fall down. It treats regulation as arbitrary friction imposed on otherwise productive companies. But a lot of what gets called “bureaucracy” is actually worker protection, consumer protection, food safety, environmental standards, privacy rights or product safety. You can call those things burdens if you’re a company. From the perspective of a citizen, they’re often the things that make a market worth trusting.

Worker protections are not pointless bureaucracy

Take labour law. Many European countries have stronger worker protections than the US. Employees tend to get more holiday. There are often stronger rules around redundancy, sick leave, parental leave, minimum wage, notice periods and dismissal. It can also be harder to fire people once they’ve been with you for a while. I can see why some founders find that frustrating, especially if they’ve been exposed to a more American model, where it’s often much easier to hire quickly, fire quickly and restructure quickly.

Still, I struggle to believe this is the thing holding Europe back. An extra few weeks of annual leave is not the reason we don’t have more trillion-dollar technology companies. A more flexible labour market gives companies more room to move fast. A more protective labour market gives workers more security. You can argue about where the balance should sit. But it’s misleading to talk about worker protections as though they’re just pointless friction.

They exist because employers have power. They were hard won over generations. They stop people being discarded casually, arbitrarily or abusively. That may be less convenient for companies. It doesn’t make it irrational.

GDPR is annoying. That doesn’t make it the villain.

GDPR is probably the example people reach for most often. And yes, GDPR is a pain. Nobody enjoys cookie banners, consent flows, data-processing agreements, vendor reviews, subject-access requests or the general swamp of privacy compliance. But GDPR didn’t appear because European regulators were bored. It appeared because companies were collecting, sharing, profiling and exploiting personal data at industrial scale.

It is also worth getting the chronology right. Amazon was founded in 1994, Google in 1998, Facebook in 2004, and GDPR came into force in 2018. So it is hard to argue that GDPR stopped Europe producing its own Google, Amazon or Meta. Those companies were created long before GDPR existed. If anything, GDPR was partly a response to the world those companies helped create: a world in which a small number of platforms had accumulated vast amounts of personal data, extraordinary market power, and increasingly sophisticated ways to track, target and influence people at scale.

What founders often describe as “data-driven innovation” can, from another angle, look a lot like surveillance capitalism: collecting far too much information about people, combining it with data from other sources, and using it in ways most people would never meaningfully understand or agree to. Shadow profiles. A contact list being scraped to spam your friends without your permission. Apps quietly sharing location data. Street View cars collecting data from open Wi-Fi networks as they drove past people’s homes. Job sites using credit-history data to infer who might accept lower pay. Companies discovering that if they know enough about you, they can predict, target, influence, segment or exploit you in increasingly intimate ways.

Cambridge Analytica became the famous example, but it was hardly the only one. The broader issue was a massive transfer of power from citizens to platforms, advertisers, data brokers and political operators. Some of this may have looked like innovation from inside the company. From the outside, it often looked like exploitation dressed up as optimisation.

This is where the usual argument gets things backwards. The problem is not that privacy rules strangled a generation of would-be European tech giants. By the time those rules arrived, the big platform companies were already powerful enough to absorb the compliance burden. Regulation introduced late can end up protecting the very incumbents it was meant to discipline, because Google, Meta or Amazon can afford armies of lawyers, policy teams and compliance specialists. A smaller competitor cannot.

So yes, GDPR may create costs for startups. It may make some kinds of data-heavy experimentation harder. It may hit smaller companies disproportionately. But that is a different argument from saying GDPR caused Europe’s lack of tech giants. In many ways, GDPR arrived after the giants had already been built.

GDPR mostly makes it harder to casually misuse personal data. It forces companies to think about consent, retention, deletion, portability, security and legitimate use. Those don’t seem like insane things to care about. The European model is more willing to ask what harms might follow from giving companies effectively unlimited access to people’s personal lives. The aim is to balance commercial freedom against the rights of citizens not to be invisibly tracked, profiled, manipulated, discriminated against or exposed.

GDPR also applies to American companies operating in Europe. So it’s hard to argue that it uniquely explains why European companies underperform American ones. It may have been badly implemented in places. It may need simplifying in others. But as a grand explanation for why Europe doesn’t have as many technology giants as the US, it feels weak. And if the only way your business works is by quietly hoovering up dubiously acquired personal data, cross-referencing it with other datasets, and making it hard for people to understand or object, maybe the problem is the business model.

Europe isn’t America, because Europe isn’t a country

The much bigger issue is also the most obvious one. America is one country. Europe is not. The US has one huge domestic market, one dominant language, deep capital markets, a federal system, a mature venture ecosystem, and a relatively coherent commercial culture. Europe is a continent of different countries with different languages, tax systems, legal systems, employment norms, public institutions, buyer behaviours and commercial cultures.

Selling from California to New York is not the same as selling from France to Germany, or from Germany to Spain, or from the UK into the EU after Brexit. Of course that creates friction. But that isn’t Europe burdening itself with pointless regulation. That’s what happens when you trade across borders. There are frictions between the US and Canada. There are frictions between the US and Mexico. Different countries have different systems. That’s normal international trade.

If anything, one of the main reasons the EU exists is to reduce this friction: harmonising standards, recognising rules, removing barriers, creating common frameworks and making it easier to do business across the continent than it otherwise would be. So there’s something odd about blaming the EU for the complexity it was partly created to solve. Europe is harder to scale across because Europe is not a single nation-state. That’s a structural fact, not a morality tale about red tape.

The real comparison is not America versus Europe

There is another sleight of hand in the usual comparison. People talk as if America is a better place to start a company than Europe. But that is rarely what they really mean. The more honest version is that San Francisco, and maybe New York, are better places to scale certain kinds of technology company than Lisbon, Copenhagen, Berlin, Paris or London.

That is a narrower claim, and a much more plausible one. The American startup ecosystem is not evenly distributed across the whole country. A founder in Kansas or Ohio does not have the same access to capital, talent, networks, early adopters and repeat operators as a founder in San Francisco. The US advantage is real, but much of it is concentrated in a handful of places. In venture, especially, it can feel as though an absurd amount of the world’s risk capital is clustered around one hilly road and one small square in San Francisco, rather than spread across dozens of national capitals.

This is one reason so many European companies start here, then move some or all of their centre of gravity to the US. It is rarely because they were crushed by European regulation. More often, it is because the US is where the later-stage capital is, where the biggest software buyers are, where the acquirers are, where the category analysts are, where the talent has already scaled something similar, and where the next round of investors expect you to be.

There is also a cultural element. American companies are often more willing to try new software if the upside looks big enough. European buyers, especially in larger or more traditional organisations, can be more focused on downside risk: what might break, who might object, whether procurement will approve it, whether legal is comfortable, whether the vendor will still be around in three years. That caution is not irrational. Sometimes it is responsible governance. But it does make it harder for new products to break in quickly.

That same difference probably sits underneath a lot of regulation too. Europe does not create stronger protections because it has an abstract love of paperwork. It tends to regulate because it has a lower tolerance for companies privatising the upside while socialising the harm. That is a different moral and political settlement, not simply an administrative defect.

Some of this may come from the texture of European life. European cities are often denser. People are more likely to live near, walk past, sit beside, or share public services with the people affected by corporate behaviour. Harm is harder to abstract away when it happens on your street, in your hospital, on your train, in your school, or to someone your family might plausibly know.

In parts of the US, especially in wealthy technology suburbs, harm can feel more spatially distant. If you live in a nice house in Palo Alto, travel to campus on a private shuttle, work in a highly paid bubble, and mostly interact with other people in the same industry, it is probably easier to see your company’s externalities as abstractions. The gig worker, the warehouse worker, the person priced out of their neighbourhood, the user being profiled, the community dealing with pollution, the family navigating medical debt: they can all become edge cases in a dashboard.

This does not mean people in tech are bad people. It means incentives and environments shape what we notice. If the system pays you well not to notice certain harms, and your daily life keeps you physically and socially distant from those harms, it becomes easier to believe that any constraint on your company is irrational bureaucracy.

Europe’s instinct, at its best, is different. It is more willing to ask who bears the downside when companies move fast. That can make it slower and more cautious. Sometimes it leads to clumsy rules. But it also reflects a society that is less comfortable letting private companies capture the benefits while leaving citizens to absorb the costs.

So when people say European regulation holds back innovation, it is worth asking what kind of innovation they mean, and whose risk tolerance they are talking about. Founders and investors may be happy to take risks. Workers, consumers, neighbours, patients and citizens may not have agreed to be part of the experiment.

Higher standards are not anti-business rules

A lot of European regulation concerns physical products: food, electronics, toys, cosmetics, medicines, chemicals, building materials, packaging and safety standards. Again, this often gets described as bureaucracy. But most people quite like knowing that the food they buy is safe, that children’s toys aren’t toxic, that electronics won’t burst into flames, that cosmetics have been tested, and that manufacturers can’t simply dump risk onto consumers.

This is one reason trade deals with America often make Europeans nervous. The debate around chlorinated chicken was never really just about chicken. It became a symbol of something larger: the fear that “market access” would become a polite way of saying, “please lower your food and safety standards so our companies can sell more easily into your market.” You can dismiss that as protectionism if you like. From a European perspective, it often looks like defending standards that citizens broadly trust.

And when Europeans look at some parts of the American system, the idea that the US has found a superior low-friction model isn’t always obvious.

America has plenty of friction. It just puts it somewhere else.

The US is not a frictionless business paradise. It has a wildly complex tax code. It has federal, state and local rules. It has state-by-state employment variation. It has sales tax complexity. It has immigration constraints. It has huge litigation risk. It has expensive lawyers. It has class actions, discovery processes, liability exposure, insurance complexity, securities law and an increasingly messy patchwork of state privacy rules. That doesn’t sound especially low-bureaucracy to me.

Healthcare is the obvious example. In the UK, if you hire someone, you hire them. The NHS exists in the background. You don’t need to design a healthcare benefits system, negotiate insurance plans, worry about networks and deductibles, manage brokers and renewals, or decide what level of medical coverage your employees and their families should receive. In the US, healthcare becomes an employer-side operational burden. Companies spend time, money and management attention on something that, in many European countries, is simply part of the social infrastructure.

That is friction. It just doesn’t always get counted as regulation because it’s administered through private systems. The same is true of litigation. If Europe has bureaucrats, America has lawyers. Europe often creates upfront rules to prevent harm. America often tolerates more risk, then deals with some of the consequences through lawsuits, insurance, private contracts, medical bills and individual exposure. Different systems. Different costs.

Low regulation can also mean regulatory capture

There’s another part of this that often gets missed. The US doesn’t necessarily have less regulation because it has discovered a more efficient form of capitalism. In some sectors, it has less protection because powerful industries have successfully lobbied to shape the rules in their favour. That can feel wonderfully low-friction if you’re the company benefiting from it. But low friction for whom?

A chemical company may prefer weaker environmental rules. An employer may prefer weaker labour protections. A food producer may prefer looser standards. A healthcare company may prefer a system so complex that patients, employers and sometimes even doctors struggle to understand what anything actually costs. A dominant tech platform may prefer privacy rules weak enough to allow near-limitless data extraction. From the company’s point of view, fewer constraints can look like efficiency. From everyone else’s point of view, it may simply mean the costs have been pushed somewhere else: onto workers, consumers, patients, communities, rivers, public health or future generations.

This is the bit of the “Europe is over-regulated” argument that often feels most suspect. It assumes that lighter rules are naturally better, without asking whether those lighter rules are the result of democratic wisdom or successful lobbying by wealthy vested interests. Regulatory capture can masquerade as dynamism. A sector can look lean and innovative because the public has quietly absorbed the downside. Pollution becomes a community problem. Medical complexity becomes an employer and household problem. Weak labour protection becomes a worker problem. Data extraction becomes a citizen problem. Unsafe products become a litigation problem. That isn’t the absence of bureaucracy. It’s the privatisation of harm.

Silicon Valley wasn’t created by low regulation alone

None of this means Europe is secretly outperforming America. It clearly isn’t, at least not in technology. But the reasons are much deeper than “too much regulation.” The US had extraordinary starting conditions: post-war industrial strength, defence spending, Cold War research funding, elite universities, immigration, deep capital markets, the rise of semiconductors, the emergence of Silicon Valley, strong links between government, academia and industry, a huge domestic market, an aggressive venture-capital ecosystem, stock options, public markets willing to reward growth, and a culture more comfortable with founder ambition and monopoly-scale outcomes.

Silicon Valley didn’t appear because California had fewer cookie banners. It emerged through generations of compounding advantage. The silicon part of Silicon Valley led to hardware. Hardware led to software. Software led to the internet. The internet led to mobile. Mobile led to cloud. Cloud led to AI. Each wave produced companies, capital, operators, angels, acquirers, infrastructure and ambition for the next wave.

Europe has strong companies. It has world-class research. It has excellent technical talent. It has produced important businesses. But it hasn’t had the same flywheel, at the same scale, for the same length of time. That’s largely a story about capital, institutions, market size, history and compounding.

Europe has real problems. They’re just not always the problems people point to.

If we want to talk seriously about Europe’s underperformance, there is plenty to discuss. Europe has less late-stage capital. Its pension funds and institutional investors have historically been less exposed to venture. Its public markets are less attractive for high-growth technology companies. Stock-option treatment is often worse. Founder upside can be weaker. Public procurement is less aggressive. Universities are not always connected to company formation in the same way. There are fewer giant domestic technology buyers. There are fewer repeat founders and scale-up executives who have been through the full journey.

Europe also suffers from fragmentation. Not just legal fragmentation, but cultural, linguistic, financial and commercial fragmentation. A startup in the US can often build for one large home market before expanding internationally. A European startup may have to think internationally much earlier, but without the same unified base from which to scale.

These are real issues. They deserve attention. Some bureaucracy really is pointless. Some permitting processes are too slow. Some compliance obligations are badly designed. Some public-sector procurement systems are painful. Some national rules create unnecessary duplication. Some European countries make company formation, equity compensation, hiring or administration more difficult than they need to be. But this is a much more specific argument than “Europe regulates too much.” Specificity matters. If you misdiagnose the problem, you reach for the wrong cure.

The danger of the lazy regulation story

The lazy story goes something like this: America builds. Europe regulates. It’s catchy. It’s just not very serious. It implies that Europe’s main problem is that it protects people too much. Too much privacy. Too much holiday. Too much food safety. Too much consumer protection. Too many limits on what employers can do.

But perhaps those things are features of the European model rather than bugs. Europe should make it easier to build ambitious companies while preserving the parts of that model that are worth defending. That means deeper capital markets, better founder incentives, more effective stock-option schemes, more pension-fund capital flowing into European venture rather than defaulting into the S&P 500, faster and smarter public procurement, stronger university spin-out pathways, better late-stage funding, more ambitious industrial strategy, a simpler route to forming and operating a genuinely pan-European company, easier cross-border scaling, less duplication, faster permitting for genuinely strategic infrastructure, and more willingness to back European champions. It does not need to mean weaker food standards, fewer holidays, worse privacy rights or making it easier to fire people on a whim.

Regulation needs maintenance, not bonfires

The goal should be regulatory maintenance, not deregulation for its own sake. Like code, regulation accumulates over time. Some of it is elegant and essential. Some of it is legacy cruft. Some of it was written for a world that no longer exists. Some of it solves a problem that has since changed shape. Some of it protects incumbents under the guise of protecting the public.

The answer to bad code is not to delete the whole system and hope for the best. It’s to refactor it. That means asking what each rule is for, whether it still works, who benefits from it, who bears the cost, and whether there is a simpler way to achieve the same public good.

A serious competitiveness agenda would distinguish between useful protections that improve life for workers, consumers and citizens; pointless bureaucracy that creates paperwork without meaningful public benefit; legacy complexity that made sense once but no longer fits the world we live in; and structural fragmentation that comes from Europe being a continent of different countries rather than one unified nation-state. Lumping all of these together under “too much regulation” is branding pretending to be analysis.

Europe needs reform. But not self-loathing.

Europe absolutely needs to get better at building. It needs more ambition, more capital, more urgency, more technical depth, more institutional competence, more comfort with scale, and more willingness to fund and buy from its own emerging technology companies. But it should be careful not to confuse reform with imitation.

The American model has produced extraordinary companies. It has also produced extreme inequality, fragile healthcare access, weaker worker power, high litigation costs, corporate capture, and business models that too often externalise harm onto users, workers and communities. That is not a free lunch.

Europe’s task is to build a better innovation machine on European terms. That means being honest about what is genuinely holding companies back, and equally honest about what is worth protecting. Pointless bureaucracy should go. Fragmented systems should be harmonised. Slow institutions should be fixed. Capital should move more easily. Founders should have better incentives. Public institutions should become better customers. Strategic infrastructure should not take decades to approve.

But worker rights, privacy, food safety, product standards and consumer protection are not the reason Europe lacks a Google. They are part of the society Europe has chosen to build. The hard problem is how to build world-class companies without pretending those protections are the enemy.

Where I think the post goes wrong is in its explanation of why some designers fall short, and what that says about them as people.

Too often, behaviours that are really about anxiety, lack of context, or organisational dysfunction get misread as arrogance, laziness, or ideological purity. And when product leaders, founders, or engineers start from that kind of mental model, it is no surprise that relationships with design become strained. If you go into a cross-functional relationship already assuming bad intent, you will tend to find evidence for it everywhere.

So yes, I think there is something useful in the four-type framework. But I think the diagnosis needs a lot more nuance.

Type 1: The anxious, overwhelmed designer

The original post describes Type 1 as someone who cannot produce something reasonable on time.

That person definitely exists. But I think the behaviour is often badly misread.

In my experience, these designers are rarely arrogant. More often they are conscientious, pretty good, and trying hard, but they know they are not quite cutting the mustard. They can feel the gap between where they are and where they need to be. That gap creates anxiety. They feel stuck, under-confident, and very wary of being exposed.

So they hide.

They need time alone to think, explore, and work things through. They disappear into the problem for days or weeks, avoid sharing work in progress, resist people looking in their files, and then emerge just before the big review with what they believe is the answer. Partly this is about craft. But often it is also about fear. They are hoping that if they can just get it right in private, nobody will notice the imposter they suspect themselves to be.

From a leadership point of view, this can be incredibly frustrating. The work has been created in isolation, with very little chance to sense-check the direction, and by the time it appears, it often does not quite meet the brief. Sometimes it is thoughtful. Sometimes parts of it are strong. But it has not been shaped collaboratively enough to be truly useful.

I think part of the problem is that these designers often see themselves as delivering an output rather than providing a service. Good product design is collaborative. You need to show your thinking early, invite challenge, and let the work improve through discussion. But if you are deeply anxious, that can feel dangerous. Early feedback feels exposing. Criticism feels personal. So instead of opening the work up while it is still rough, they hide away until they feel certain it is good enough.

And that is where the pattern becomes self-defeating. The certainty is often false. So when the feedback finally comes and the work is only 60 or 70 percent there, what gets read as arrogance or spikiness is often just defensiveness. “You don’t understand the design process.” “Leave me alone.” “Stop seagulling.” Underneath that is often not ego, but fear. A designer trying to protect themselves from feeling found out.

That does not make the behaviour any less problematic. It can create drag, waste time, and make collaboration much harder than it needs to be. But I think it is better understood as fear than arrogance.

The tragedy is that, in the current environment, this kind of designer is particularly exposed. AI can already generate decent-looking work at astonishing speed. So if your value is producing a 90% answer in ten times longer than the machine, you are in trouble. The market is becoming much less forgiving of slow, craft-heavy, low-judgement design.

That is uncomfortable, but I think it is true.

Type 2: The principled designer who gets read as a blocker

The original post describes Type 2 as someone who designs well, but refuses to understand the business because they think it is beneath them.

Again, I think there is a real behaviour being pointed at here, but the motivation is often misread.

There are certainly designers who over-index on user advocacy in a way that becomes impractical. They ask hard questions about every initiative. They challenge assumptions. They worry about edge cases. They point out inconsistencies. They raise concerns about whether a feature is actually needed, or whether it introduces confusion, friction, or harm.

Sometimes they are right.

Sometimes they are exhausting.

The problem is that these behaviours can be interpreted in very different ways depending on where you are sitting.

From the designer’s perspective, they are being careful, conscientious, and principled. They are trying to protect the user, the product, and in many cases the company from bad decisions.

From the product manager’s perspective, especially under pressure, it can feel like death by a thousand cuts. Every idea gets questioned. Every plan gets slowed down. Every decision turns into a philosophical debate. What feels like integrity on one side can feel like obstruction on the other.

And because many PMs are already under pressure from leadership to move fast and deliver outcomes, they do not always experience this kind of design pushback as helpful rigour. They experience it as challenge, friction, and sometimes ego.

I do think some designers in this camp have not fully internalised the commercial reality of the environment they are working in. They may feel a strong moral duty to the user without quite appreciating that, if the business does not succeed, they will not get to help users for very long. They may treat any compromise as a betrayal rather than what it often is: a trade-off.

But “thinks business is beneath them” is, I think, a lazy reading.

A lot of these designers do not look down on the business. They simply care most naturally and fluently about the human side of the work. That is not a flaw. It only becomes a problem when it is disconnected from commercial reality.

Type 3: The designer trying to care about the business without enough context

The original post describes Type 3 as the designer who says they care about the business, but does not really understand it and is, to some degree, lying to themselves.

This is where I think the critique becomes particularly unfair.

There are designers who use the language of business fluency a little too casually. They know the terms. They can talk about strategy, growth, conversion and retention. But if you scratch below the surface, the understanding is not always deep.

That is real.

But I think the bigger issue is not insincerity. It is lack of exposure.

Many product managers and business leaders are in rooms that designers never get invited into. They hear the nuances of the commercial conversation. They understand the trade-offs leadership is wrestling with. They know which metrics matter, which customer segments are strategic, where margins are thin, where support costs are rising, where churn is hurting, and what the market is doing.

Then that rich, messy context gets compressed into a roadmap.

Then the roadmap gets turned into a PRD.

And by the time it reaches design, much of the underlying rationale has been flattened into: “We need this feature” or “Please improve this flow.”

At that point, the designer is being asked to make product decisions without being given the material that would help them make those decisions well.

So yes, many designers want to be more commercially minded, but are not yet particularly good at it. Not because they are pretending. Not because they secretly think the business is beneath them. But because they have never really been taught how the machine works.

And because they do have strong instincts around user experience, product quality, or what feels coherent and compelling, they sometimes push back in ways that are hard for the business to understand. Especially when the business case has not been clearly shared.

This is one of the great sources of tension in modern product teams. Designers are asked to think strategically, but are often denied strategic context. Then they get criticised for not showing enough strategic judgement.

Type 4: The designer every company wants

The original post’s Type 4 is the designer who designs well, understands the business, and actively spots opportunities to improve both the user experience and the economics of the company.

This person absolutely exists.

They are fast without being sloppy. They understand that the goal is not perfection, but progress. They know when to explore and when to converge. They care about users, but they also care about activation, conversion, retention, support load, operational efficiency, and growth. They can connect a frustrating workflow to support costs. They can connect a clunky onboarding journey to time-to-value. They can connect a confusing plan page to sales friction or poor self-serve conversion.

They do not just make screens better. They make decisions better.

These designers are hugely valuable because they reduce the translation gap between design, product, engineering, and the business. They can hold all those considerations in their head at once and move fluidly between them.

I agree completely that companies need more of these people.

Where I disagree is the implication that everyone else should simply be written off.

Because I think there are a lot more potential Type 4s in the system than people realise.

The real question: how do Type 2s and Type 3s become Type 4s?

This is where I think the original argument is weakest.

It treats designer capability as if it were mostly fixed, based on character, and as if the problem is simply hiring the wrong people. Hire more Type 4s. Avoid Type 2s. Get rid of the blockers. Problem solved.

Real organisations do not work like that.

In reality, many Type 2 and Type 3 designers are not static categories. They are transitional ones.

There are designers with strong craft and strong user instincts who could become commercially fluent if given the chance to see how decisions get made.

There are designers who could become faster and more decisive, but only if they get coaching on scope, prioritisation, and how to recognise when “good enough” is good enough.

There are designers who could become excellent product partners, but only if product managers stop treating them as executors and start treating them as thinkers.

If all a designer ever gets is a stream of poorly thought-through PRDs and a request to “make it look good”, then of course they will struggle to develop product judgement. You cannot ask people to think like product leaders while structuring their role like production support.

A lot of the designers companies say they want are never actually given the environment required to become those designers.

Why this matters beyond design

This is not just a design critique. It is a cross-functional one.

Because I strongly suspect that part of the tension between designers, product managers, and engineers comes from these misdiagnosed mental models.

If product peers see designers as precious, anti-business, slow, or vaguely self-righteous, they will tend to interpret their behaviour through that lens. A thoughtful question becomes resistance. A request for context becomes politics. A concern about usability becomes naivety. A desire to explore becomes indulgence.

Likewise, if designers see PMs as shallow feature factories or engineers as people who only care about shipping tickets, they will misread motivations too.

The result is predictable. Everyone feels misunderstood. Everyone becomes more defensive. The relationship gets more transactional. Trust goes down. Collaboration worsens. And each side walks away feeling confirmed in its worst assumptions about the other.

That is why I think language like the original post is not just abrasive. It is actively counterproductive. It may feel cathartic to people who have had poor experiences with certain designers. But it hardens caricatures at exactly the moment when most teams need more empathy and better translation across disciplines.

So where do I land?

I think there is real truth in the claim that the highest-value designers are not just stylists or user advocates. They are people with product judgement, commercial awareness, and the ability to move at pace.

I think there are also designers who are too slow, too perfectionistic, too detached from commercial reality, or too vague in their understanding of the business.

But I do not think the right response is contempt.

Most of these behaviours are not signs of bad character. They are signs of incomplete development, poor context, a lack of support, and of organisations that have not figured out how to integrate design well.

The goal should not just be to identify the rare mythical Type 4.

It should be to create more of them.

And that means giving designers more exposure to business context, more feedback on trade-offs, more support in building product judgement, and more trust to operate as real partners rather than downstream executors.

There is a useful framework here.

It just needs a lot less sneering, and a lot more understanding.

Once that happens, the economics of thought begin to change. It is often quicker to ship an AI-generated prototype than it is to run a considered design process. And if the prototype is wrong, the instinct is no longer to avoid the mistake in the first place. It is to fix it tomorrow, or ship another version by the end of the week. Teams are no longer trying to play the best, most considered game of chess, where the best prepared player wins. They are trying to speed up the play of poker, getting through as many hands as possible, placing lots of small bets, and staying in the game long enough to get dealt pocket aces.

That is a very different philosophy of winning. Poker is faster, messier and more fluid. You make decisions with incomplete information, read the table, act under uncertainty and adjust as new cards appear. You do not wait for certainty because certainty never comes. What matters is not whether you have thought everything through, but whether you can move quickly enough and keep playing long enough for the odds to work in your favour. That feels much closer to how many product teams operate now.

As that shift took hold, process started to collapse. Over the last decade, the combination of ever-growing backlogs, shorter delivery cycles and relentless pressure to ship has squeezed out many of the conditions thoughtful design once depended on. Not all at once, and not in some dramatic moment, but steadily and almost invisibly.

Wireframes and paper prototypes we seen as wasteful. Especially when so many designers felt they could achieve more, faster and at a higher fidelity in Figma. Design thinking was lampooned. The Double Diamond was pronounced dead on arrival. Personas were considered harmful. Good formative research became poor customer discovery. Usability testing got replaced by build, measure, learn. We even argued about what to call ourselves. Everything we used to apply rigour, slow things down and switch into System 2 thinking got stripped out. And what was left? Delivering PRDs from somebody else's backlog.

Design did not disappear so much as get compressed, hollowed out, industrialised. Turned from a mode of inquiry into a mode of delivery. Work became less about framing the problem, exploring alternatives, or surfacing hidden assumptions, and more about keeping pace with the machinery of shipping.

That is what creates the strange irony of the current moment. Everybody says they want more design thinking, better product judgment, better user empathy and better strategic choices. Hell I've even seen folks claim that designers are needed more than ever before. They just do not want to give designers the time or space to think in the ways that used to produce those outcomes. What they want is the output of slow thinking without the slow thinking bit. They want the confidence that comes from research without waiting for research, and the answer without the wandering. They want the outcomes of design thinking, but not the tools designers once used to get there: the workshops, the sketching, the divergent exploration, the rough prototypes, the user interviews, the whiteboard sessions and the false starts.

Teams still need comprehension. In many cases they need it more than ever. But they no longer have much patience for the visible rituals that used to create it. That matters because the real bottleneck in many fast-moving teams is no longer production. It is comprehension. The difficulty is not making things. It is understanding what is happening well enough to make good decisions: understanding the user, the market, the trade-offs, the second-order effects, the hidden risks, and the reasons something feels off before the metrics tell you so. Production has become so fast that comprehension is now the thing that feels slow, and slow things always come under pressure.

So decision-making gets relocated elsewhere. It moves into product instinct, founder conviction, engineering constraints, executive preference, market momentum, or vibe-based calls made in the moment by whoever has the strongest opinions and the least patience. This is one of the more important shifts in modern product development. People still say they value design, but often what they mean is that they value taste and vibes rather than thought and process. They want someone who can look at a screen and instantly say, yes, this feels right. They want someone who can steer the work in real time. They want judgment on demand. They want fast intuition, not slower reasoning. Because taste is instantaneous, vibes are instantaneous, and thought is not.

When it is quicker to generate and ship a prototype than to run a considered design process, thought starts to look like friction. That is why the current moment feels so exposing. AI has arrived in an environment where process was already weakened, where much of design had already been recast as a delivery function, and where decisions had already started moving upstream or sideways. The new tools simply make that reality harder to ignore. If a team mainly values rapid production, quick iteration and surface-level polish, then AI is going to look increasingly competent. And when that happens we start to reach an event horizon.

The harder parts of design have not gone away. Framing the right problem still matters. Understanding human behaviour still matters. Making sense of messy situations still matters. Knowing which bets are worth placing still matters. But those forms of value are slower, less theatrical and harder to see in a culture obsessed with velocity.

That may be one reason we are drifting back toward a world of genius design. Not in the romantic sense of the lone visionary, but in the older sense that the work depends on a small number of people with highly internalised pattern recognition making rapid calls under pressure. What product teams now describe as product sense can sometimes be exactly that: design thinking compressed, internalised and turned into instinct. Sometimes that instinct is real expertise. Sometimes it is just vibes masquerading as judgment. And when process disappears from view, it becomes much harder to tell the difference.

That is the risk with vibe-based decisions. They feel logical in the moment. All System 1 decisions do. They create momentum. They sound confident. They help teams move. But they may also be sending the product quickly in the wrong direction. The faster the cycle time, the easier it is to mistake motion for insight.

This is the tension underneath all the current talk about AI and design. The question is not whether AI can generate artefacts. Of course it can. The more important question is whether organisations still care enough about thought to make room for it. Because if they do not, what survives will be a thinner version of design: styling at speed, prompted production with a layer of taste, and fast bets guided by instinct rather than careful reasoning.

Maybe that is enough for some teams. Maybe, in some contexts, it is even the right trade-off. But we should at least be honest about what is happening. Product development now rewards compressed cycles, quick bets, fast judgment and plausible output. It rewards enough taste to make things feel coherent, but often not enough patience to understand whether the underlying decision was wise.

That leaves designers with a genuine challenge. The task is not to defend every old ritual as sacred, nor to pretend we can rewind to a slower era. It is to work out what thoughtful design looks like in a world where process has collapsed, System 1 is crowding out System 2, and it is often faster to ship something half-right than to think something through properly.

That world is not going away. The tools will keep getting faster, and the distance between idea and execution will continue to shrink. Unless designers find ways to reintroduce comprehension, reflection and judgment into that environment, design will continue to be reduced to vibes, not in one dramatic moment, but one fast decision after another.