Lions Law: What Happens When AI Can Make Anything — but Brands Still Need to Mean Something

AI has made content dramatically easier to produce.

That is useful. It is also creating a new problem.

When everyone can generate polished words, images and video in seconds, polish stops being the advantage. A brand can produce more content than ever and still become less distinctive.

That is the problem behind Lions Law.

Lions Law is a framework I developed for governed brand execution. At its heart is a simple principle:

The idea has to live inside the execution itself. It cannot be added afterwards as an explanation.

A beautiful image with a clever caption pasted underneath is not automatically a strong idea. A video that looks expensive but could belong to ten competitors is not automatically good brand work. And an AI system producing endless variations of the same familiar category scene is not creativity simply because every version is technically competent.

The execution itself needs to carry the thinking.

Why AI content all looks the same: the danger of competent sameness

This matters more now because generative AI is exceptionally good at recognising patterns.

Ask for advertising for a lawyer and you may get polished offices, documents and serious-looking professionals. Ask for a coffee brand and the machine already understands cups, beans, steam and cafés. Ask for financial services and it knows the visual language of graphs, screens and confident businesspeople.

None of those things is necessarily wrong.

But recognising the category is not the same as having an idea.

The danger is competent sameness: work that looks appropriate, reads correctly and passes every superficial quality test, while contributing almost nothing new to the brand.

Lions Law asks the system to go further.

• What is the human truth?
• What convention is the category repeating?
• What does this particular brand have the authority to say?
• What is the organising idea?

And then: has that idea genuinely made it into the execution?

Strategy should survive the journey

Traditional brand work already has a version of this problem. A strategy can be excellent in the presentation and disappear completely by the time the advertisement, social post, video or website is produced.

AI multiplies that risk because there may now be many machines involved.

One system develops the strategy. Another writes. Another generates imagery. Another edits video. Another publishes. Another studies performance and proposes what should happen next.

Without governance, every stage gets an opportunity to reinterpret the brand.

Lions Law treats the organising idea as protected creative state.

The sequence is deliberate:

brand truth → strategy → organising idea → execution

The later stages are allowed to interpret the idea for their medium. They are not automatically allowed to replace it.

A 15-second video should not be identical to a print advertisement. A social post should not behave like an outdoor billboard. Audio needs to work through sound. Packaging occupies a physical world.

Variation is necessary.

Loss of meaning is not.

This is the same principle behind how TheAgencyIQ runs brands: each brand keeps its own purpose, voice and visual identity, enforced by the platform rather than by a template with the logo swapped.

Brand consistency now has a second audience: machines

Brands are no longer communicating only with people.

Search engines, LLMs, recommendation systems, answer engines and autonomous agents increasingly interpret businesses too.

They decide what a company appears to do, what it is associated with, whether it looks authoritative and, increasingly, whether it is worth recommending.

That means brand consistency is developing a second dimension.

It is no longer enough for a person to recognise the business correctly. Machines need to form the right understanding too.

Lions Law therefore goes beyond creative consistency. It is about maintaining authoritative brand meaning while allowing that meaning to travel through different human and machine touchpoints.

The expression may change.

The underlying truth should not quietly mutate.

Learning needs governance too

There is another trap.

Once AI systems can observe performance, it is tempting to let them continuously optimise the brand.

But evidence and authority are different things.

A post getting more engagement does not necessarily mean the positioning should change. A machine misunderstanding the company does not make the misunderstanding correct. A publishing success tells us something about delivery; it does not automatically tell us what the brand should become.

A governed system should be able to learn without allowing every signal to rewrite its foundations.

The better loop is:

observe → evidence → propose → verify → authorised change

That distinction is going to matter enormously as autonomous systems become more capable. It is why every published post in TheAgencyIQ is recorded as a verifiable outcome rather than a dashboard promise — see the measurement record.

The future advantage is not better generation

Most companies will eventually have access to very capable models.

The scarce thing will not be the ability to generate.

It will be the ability to maintain meaning while generating at scale.

That means knowing what is true, what may change, what must remain protected, who has the authority to change it, and whether the execution still carries the idea that made the brand distinctive in the first place.

That is what Lions Law is designed to govern.

Because when machines can make almost anything, the question becomes much more interesting:

Can they make something that is unmistakably, meaningfully and accountably yours?

Frequently asked questions

Why does AI-generated content all look the same?

Because generative models are pattern recognisers. Asked for a category, they return the category's most common conventions — the offices, the cups, the graphs. The output is competent and recognisable, which is exactly why it is not distinctive. Recognising the category is not the same as having an idea.

What is Lions Law?

A framework for governed brand execution. Its central rule is that the organising idea must live inside the execution rather than be explained afterwards, and that the idea is protected state as work travels from brand truth through strategy to the finished post, video or page.

How do you keep a brand consistent when AI is producing the work?

Fix the sequence and protect the idea across it: brand truth, strategy, organising idea, execution. Later stages may interpret the idea for their medium — a 15-second video is not a print advertisement — but they may not replace it. Consistency is about meaning surviving the journey, not every asset looking identical.

Should AI be allowed to change brand positioning based on performance?

No. Evidence and authority are different things. Engagement tells you something about delivery, not about what the brand should become. A governed loop — observe, evidence, propose, verify, authorised change — lets a system learn without letting any single signal rewrite its foundations.

Related reading: AI Made Content Cheap. Here's What Got Expensive. and Your Clients Are Worried About How You Use AI. For agencies running this across a client roster, see Qy Agency Partners.