After months of debate and final negotiations that stretched long into the night, EU officials were jubilant when, as 2023 came to a close, the European Parliament reached an agreement on what would become the EU’s AI Act; a sprawling legal framework designed to regulate artificial intelligence across all member states.

The EU’s top officials and lawmakers were delighted with their effort, and images capturing the moment were released. Almost immediately, social media posts began contrasting this tremendous act of bureaucratic achievement with the creative and engineering feats of other countries, notably the US, with pictures of SpaceX launches and frontier AI lab founders held up in contrast to the EU hailing more red tape as a moment of celebration.

In Silicon Valley the memes really took off; “America builds, China scales, Europe regulates.” The EU’s law came into effect in August 2024, triggering another round of self-congratulatory images emanating from Brussels. That was nearly two years ago, and much has changed since.

While Europe’s leaders convinced themselves that they’d done the right thing in establishing the world’s strictest AI guardrails, companies in the US and China were pushing AI into a world that the EU’s Act barely touched on. Chip manufacturing, compute power, energy supply and frontier AI labs were powering the likes of OpenAI and Anthropic towards agentic AI and artificial general intelligence while the EU thought that future battles would be about workers’ rights and tough data protection rules. These are not unimportant considerations, but the EU’s instinctive focus on regulating meant they’ve missed out on the innovating.

Europe faces a dystopian AI future

Meanwhile the UK, thanks to Brexit, was spared the EU’s ‘safety first’ approach and was able to capitalise on its regulatory freedom and sector expertise to find its own advantages in the great AI race.

Last week, a team of European researchers and academics published a scenario, Europe 2031, that paints a dystopian view of a Europe first left behind and then squeezed into irrelevance by AI’s unrelenting advance and the political and financial muscle built by the technology in the US and China. The authors envisaged a scenario where, in 2029, the US cuts Europe off from US-grown AI. By executive order, Donald Trump did something very close to this just last week. It seems that reality is catching up with fiction.

The Europe 2031 paper should be read by every minister in every EU country. We should take it seriously, too. We have advantages and opportunities, many of which the EU have cut themselves off from, but we also have a huge amount of catching up to do – on compute and energy supply – if we want to keep our head above water as the AI tide comes in.

It may be too late from Europe, but it’s not too late for us.