Culture Isthe Long Game.TechnologyIs the Lever.

For most of history, intelligence was expensive. If you wanted something analysed, drafted, translated or designed, you needed a trained person and you needed their time. Whole professions rested on that scarcity. Whole countries organised their education around it.

That scarcity is ending. I use these systems every day. I build with them. I am not interested in either panic or worship, and I hold no strong view on when machines will do this or that. What I am sure of is the direction: a growing share of competent work now costs very little, and it is available to anyone with a connection.

Think about what that does to value. When something becomes abundant, its price falls and the value moves to whatever remains scarce. This is the oldest rule in economics, and it applies to intelligence as it applied to everything else. So the serious question of this period is not what the machines can do. It is what will be scarce when they can do it.

The serious question is not what the machines can do. It is what will be scarce when they can.

My answer is identity. Who you are, what you stand for, what you remember, and how that is expressed. A model can imitate a style, but it cannot have a stake in anything. It has no people, no grandmother, no reason to keep a promise.

The fourth of my laws says that culture is the long game and technology is the lever. I wrote it before this wave arrived, and I now think it matters more than the other three.

The way I frame this in my own work is as a multiplication. It is the reason Artmand has two divisions that look so unlike one another. Artmand Intelligence works on systems. Artmand Identity works on meaning. People sometimes ask why one company would hold both. My answer is that they are two terms in one formula.

Intelligence without identity is a commodity. This is the bottom of the market that is being created right now: capable, tireless and interchangeable. Any organisation that competes here is competing with the next model release, and it will lose. Companies are rushing to add AI to their products. If that is all they add, they have made themselves more similar to everybody else.

Identity without intelligence is heritage. I have enormous respect for it, and I also see the risk, because I come from it. A people can possess a magnificent culture and still become a museum of itself. It is admired, photographed, funded by foreign foundations, and slowly absent from the present. Heritage that has no leverage is a beautiful way of losing.

Heritage with no leverage is a beautiful way of losing.

The corner worth occupying is where both are strong. A known character with the power of a machine behind it. It does not matter whether we are talking about a person, a brand or a nation. It is the same corner.

Note that this is a product and not a sum. In a sum, strength in one term can make up for weakness in the other. In a product it cannot. If either term is close to zero, so is the result.

A lever is a precise image, and I chose it with care. A lever multiplies force. It does not create any. It also needs two things that it cannot provide for itself: a place to stand, and a decision about what to move.

Technology is the top layer here, and the most impressive one. It is also the most easily replaced. Beneath it sit institutions, which keep a purpose alive across many people and many years. Beneath them sits language, which sets the limits of what can be said and therefore of what can be thought. At the bottom sits memory, the things a people refuses to forget.

This is not a romantic argument. It is an engineering one. Models learn from what has been written down. A language with a thin digital record gets a thin machine. For a language like Kurdish, this is an immediate and practical matter. Whatever is not recorded, digitised and structured now will be poorly served by every system that follows, and the gap will widen.

So the most forward-looking cultural work today can look oddly old-fashioned. Archives. Dictionaries. Careful transcription. Typefaces that can set a script properly. Good publishing. This is the fulcrum. Anyone who wants the coming tools to work for their people must build it now, and nobody else will do it for them.

There is a personal version of this too. I notice it in my own studio. A tool can produce a hundred competent logos before lunch. It cannot tell you which one your client's grandmother would recognise as theirs, or which colour carries a memory that a stranger would miss. That judgement comes from having lived inside a culture and paid attention to it. So the designers who will do well are not the ones who refuse the tools, and they are not the ones who surrender to them. They are the ones who bring something the tool could not have known, and then use it to go further than they could have gone alone. The machine can extend a point of view. It has none of its own to extend.

Part of the anxiety of this moment comes from a confusion of speeds. Things that move at very different rates are being compared as though they were the same thing.

Models are replaced before most people have finished learning them. Products last a few years with care. Institutions last decades if they are well made. Culture moves slowly and decides everything.

The fast layers get all of the attention, because fast things make news. But the fast layers are shaped by the slow ones. What a society chooses to build with a new tool, what it forbids, what it finds beautiful or shameful: none of that comes from the tool. A technology arriving in a culture with weak institutions and a thin sense of itself will simply be used on that culture by somebody else.

A technology arriving in a culture with a thin sense of itself will be used on that culture by somebody else.

This is why I am suspicious of two popular postures. The first is the rush to adopt everything so as not to be left behind. The second is the refusal to touch anything so as not to be contaminated. Both of them are reactions to speed. Neither is a strategy. The strategy is to know which layer you are working on and to give it the time it actually requires.

I want the fast tools in my hands. I run them through everything I build. But I want them pointed at something that was decided slowly.

If you run an organisation, the practical conclusion is uncomfortable. The more AI you adopt, the more identity work you will need, not less. When every competitor has access to the same intelligence, your voice, your standards and your point of view are the only remaining differences. Most companies have let these drift for years, because they did not seem urgent. They are urgent now.

So write down who you are, in terms specific enough to refuse something. A value that has never cost you anything is a decoration. Decide what you will not automate, and why. Keep a human name on the things that matter. And treat your archive, meaning your documents, your language and your institutional memory, as the strategic asset it has just become.

If you lead a people or a nation, the same applies at a larger scale. Fund the slow layer. Digitise your language with seriousness. Teach your children both the old stories and the new tools, and do not let anyone persuade you to choose between them. A country that can only consume intelligence built elsewhere has accepted a quiet new form of dependency.

And if you are young and talented and wondering whether any of this leaves room for you, I think it leaves more room than ever. Machines have made competence cheap. They have not made it any cheaper to have something to say. The person who knows who they are, and who has done the slow work of becoming someone, now holds a lever more powerful than anything their grandparents could have imagined.

That is the reason for my optimism, and it is a practical optimism. The long game has not changed. We have simply been handed a longer lever.

The long game has not changed. We have simply been handed a longer lever.