The EU AI Act Will Be Enforced On Your Engineers
The compliance calendar for the EU AI Act has been updated again. Law firms are now maintaining living trackers of the key dates for compliance, which tells you something on its own. The Act does not land in one piece on one morning. It phases in, obligation by obligation, and the version of the calendar you read six months ago is not the one that governs you now.
Most of the hiring plans I see treat this as a legal matter, which is the reading I think is wrong. The Act does not mainly ask whether your model is good. It asks whether you can show your work: where the training data came from, how the system was evaluated, who is accountable for oversight, what happens when it fails. None of that can be assembled by counsel after the fact. It has to be produced while the system is being built, by the people building it, or it does not exist at all.
Compliance is not a document your lawyers write about the system. It is a property your engineers build into it.
So the repricing is happening inside technical teams rather than beside them. The scarce person is not the policy specialist and not the model builder. It is the one who can do both, who can ship something into production and then account for it in terms a regulator or an enterprise procurement team will accept. Companies that respond to this by hiring a governance lead and pointing them at the engineering org will discover the governance lead cannot answer the questions, because the answers were never written down. Meanwhile the engineers who have been documenting their systems all along quietly become the only people who can make a European launch happen on time.
For hiring leaders, the practical read is simple. Add one requirement to your next AI hire, at any level: ask the candidate to explain in writing how they would evidence a system they have already built. The ones who can do it in a page are the ones your deadline will depend on, and they are cheaper to hire before the deadline is visible than after.
Planning a senior AI or data hire and want the market view first? We share it either way.
Send us the brief →