This week's edition covers the EU's €1 billion push to actually use AI, the hardware crisis making AI less accessible, and the UK's regulatory sandbox experiment—three stories about the gap between AI policy and practice.
The European Commission has just launched its Apply AI Strategy, and this time, it's not about more rules—it's about getting AI out into the real world.
After years of regulatory enthusiasm, they've noticed everyone else is actually building things
They're putting about €1 billion on the table, spread across eleven key sectors, using programs like Horizon Europe and Digital Europe. Why? Because right now, only 13.5% of EU businesses are actually using AI.
So we've spent years perfecting the rules for a party barely anyone showed up to
That's a pretty small slice, and the Commission wants to change that.
What's the plan? Instead of just talking about AI, the EU wants to actually put it to work. In healthcare, they're setting up AI-powered screening centers to catch heart disease and cancer early, using huge European health data sets. For transport, they're building real-world testbeds for self-driving cars that can cross borders, using special regulatory 'sandboxes' to let new ideas run free.
The EU creating exemptions from its own regulations—that's character development right there
Pharma companies get a shot at an AI drug discovery challenge, with winners getting access to serious computing power. Manufacturers are being encouraged to pool their data—safely, through trusted third parties—so they can build better AI models, all while keeping their trade secrets safe.
Share everything, trust no one—the dating advice approach to industrial strategy
There's also a new 'AI first' policy, nudging organizations to pick European-made AI tools whenever they can.
Government procurement preferences disguised as innovation policy—a tale as old as protectionism itself
The old AI Alliance is getting a makeover, turning into a place where businesses and policymakers can actually talk to each other and work together. And to keep score, an AI Observatory will track how much AI is actually being used, and help set targets for public and private investment.
But here's the catch: the Commission wants every EU country to get on board with this sector-by-sector approach. The problem? Many countries haven't even set up the national authorities they need to run things under the new AI Act.
Asking countries to implement your adoption strategy when they haven't finished implementing your regulation strategy is... optimistic
And while €1 billion sounds like a lot, it's pocket change compared to what the US and China are spending.
When your "moonshot" budget wouldn't cover a decent Silicon Valley office lease
Plus, by pushing for European-made AI, the EU risks falling behind when most of the world is already using global tools.
Why should you care? If you're in one of these eleven sectors, now's the time to look for EU funding and get involved in these new testbeds—especially if you're already using European AI tools.
Both of them
The revamped AI Alliance is your chance to actually talk to policymakers and shape how things roll out. But be careful: betting everything on European AI could be risky, since the market is still dominated by global players. The smart move might be to mix and match—keep the EU happy, but don't lose your edge.
For policymakers, the real headache is getting everyone to work together. The whole plan depends on having the right systems in place, but many countries are still scrambling to build them—even though the AI Act rules kick in soon. There's not much time left, and if national priorities clash with the EU's big vision, things could get messy.
Twenty-seven countries, twenty-seven interpretations of "urgent"—what could go wrong
If you're in compliance, pay attention: there's a whole new layer of governance growing alongside the AI Act.
Because what every regulated industry was missing was more layers
The AI Observatory will be watching who's using AI and how, so there's nowhere to hide. The focus on European solutions and keeping data local means that companies building on EU infrastructure could get a leg up. But if you work across different EU countries, you'll have to juggle both the rules and these new adoption programs.
The bureaucratic equivalent of patting your head and rubbing your stomach, but with billion-euro consequences
More AI everywhere doesn’t mean better results. Small teams of AI experts beat mass licenses. Attackers now use AI to run cybercrime at machine speed. Leaders can’t keep up. Tools still forget us. The winners? Focused expertise and hybrid AI, not AI everywhere.
You've built something and you need to know if it works. So you do what's sensible—you ask an LLM to grade it. Factual accuracy, code quality, agent outputs. The machine judges the machine, and you get a number you can act on. Except that number
Imagine you have a bunch of teams, some with AI, some without, and some where everyone gets their own AI. Researchers ran a big experiment with over 400 people to see what actually happens when you mix and match humans and AI in different ways.
Here’s what they found:
