When manipulation becomes a ban

There is an AI Act article that I read as a cultural step, even before a legal one. It is Article 5(1)(a). It has applied since 2 February 2025.

It bans placing on the market, putting into service or using systems that deploy subliminal techniques beyond a person's consciousness, or purposefully manipulative or deceptive techniques. With a specific objective or effect: materially distorting the behaviour of a person or group by appreciably impairing their ability to make an informed decision. And thereby causing them to take a decision they would not otherwise have taken, in a manner that causes or is reasonably likely to cause that person, another person or group of persons significant harm.

What it really protects

The rule is full of conditions and must be read carefully. It does not ban persuasion in general. It does not ban advertising. It bans a specific combination: manipulative techniques, material distortion, significant harm.

But the protected interest is clear. It is not property, nor privacy in the strict sense. It is the ability to make an informed decision.

That is exactly the territory of Habeas Mentem. After the body and after data, the law is starting to protect the conditions in which we understand, judge and decide.

The problem of proof

One huge difficulty remains: how do you prove that a technique impaired the ability to decide?

Here the law needs new tools. Nobody's mind needs to be read. What needs measuring is the environment: how the choice is built, what the person sees, how much time they have, which alternatives remain visible. These are observable and comparable elements.

It is the work we do at Neuro Legal Cortex on legal texts: measuring understanding instead of presuming it. The same method can help assess an interface or a conversational system.

The ban exists. Now we need a way to make it enforceable without turning it into a slogan.