Brussels is asking its governments, not its AI firms, to make adoption happen

Brussels has spent three years writing rules for what AI cannot do. Its new flagship, the Cloud and AI Development Act, asks a different question: how do governments make AI happen? The design choice, putting duties on member states instead of on the firms building the technology, is the news.

Under CADA, EU countries must write national cloud and AI strategies, stand up "Centres for AI," designate at least one "data-centre acceleration zone," and strip out data bottlenecks. The Commission and member states must also guarantee that designated frontier-AI projects get the compute they need. Gabriela Zanfir-Fortuna of the Future of Privacy Forum calls it arguably the EU's first comprehensive tech regulation to impose "positive obligations" to enable adoption. Where the AI Act constrained what companies may do, CADA obligates what states must do.

The pattern is portable. When a media firm learns that three journalists with AI produce roughly what ten once did, as the EUObserver analysis frames it, the company is not making its staff twenty percent better. It is discovering that the entry-level rungs of the trade simply stop existing. The technology does not displace workers. It removes the on-ramp. CADA, by asking states to accelerate compute and adoption without paired obligations to the workers displaced, runs that experiment across a continent.

Watch whether the "Centres for AI" and acceleration zones come with rungs, or only with servers.