At TechBBQ, Europe’s AI conversations kept coming back to: Who’s actually in control?

At TechBBQ in Copenhagen, the loudest AI conversation wasn’t about what to build next. It was about who gets to hold the keys. Founders, investors, and operators kept circling the same issue across the conference floor, on stage, and long after the panels ended: Europe’s control over the systems behind AI.

That question landed hard after Anthropic’s Mythos and Fable models became unavailable to users outside Europe earlier this year. For some people in the room, it was a wake-up call about dependence on American and Chinese AI infrastructure. For others, it was more of a warning label than a crisis. One startup exec said the disruption hit his software team. Another sounded far calmer, even while admitting that leaning on two geopolitical powers for core AI could eventually create serious problems for Europe.

The conference theme, “Emerging from Agency,” fit the mood neatly. Ellen de Brever, an angel investor and head of partnerships at the Novo Nordisk Foundation Cellerator, said the discussion had shifted from what AI can do to what people are willing to let it do. Her point was simple and a little uncomfortable: the real issue isn’t smarter machines, but human judgment, and who stays in the driver’s seat. The conference wrapped up on August 27.

That tension showed up most clearly in the privacy debate. In a panel I moderated with Signal president Meredith Whittaker, she pushed back hard on AI assistants and agents being stitched into operating systems, including what ChatGPT is doing with iMessage. She called the current moment a “data collection apparatus” and argued that AI companies are using slick branding to blur the cost of gathering so much information. Whittaker said there’s still a huge market for privacy, and that sovereignty concerns could make that demand even stronger in Europe.

Emad Mostaque, co-founder of Stability AI and Intelligent Internet, took the argument further. On a panel about the economy and personal sovereignty, he described sovereignty as the ability to resist power being exerted over you. He warned about power concentrating inside a small number of AI labs, then went straight for the jugular: if AI keeps centralizing, the person controlling it controls the country. Mia Negru of Life With Artificials took that thought even further in a message after the panel, saying agentic AI and robotics could force a rethink of ownership, work, democracy, and even who gets to participate in society.

And yet, TechBBQ was not all grim geopolitics and existential debate. Lovable, Nvidia, OpenAI, AWS Startups, and HSBC hosted a rooftop barbecue. Ada Ventures threw a cocktail gathering at Bird. There was a speakers’ dinner on an island in the middle of Copenhagen, with fire dancers and an international press table, followed by an after-party upstairs where people tried to squeeze in one more drink before the bar tab closed. The big takeaway from de Brever was almost charmingly low-tech: even in a room full of talk about machines, the best moments still came from people meeting face to face.