IMPLICATOR
A speculative scenario published June 11 by a group of European AI researchers and investors puts the continent at roughly 5 percent of the world's AI computing capacity, against about 80 percent for the United States. The document, titled Europe 2031, traces a fictional five-year slide into economic and political dependence and pairs the story with real data through mid-2026, including the distance between the largest American AI supercomputer, at 1,250 megawatts, and Europe's largest, at 83. It went viral during this week's Group of Seven talks in France, read by members of the European Parliament and, its authors say, raised in unofficial British-German discussions, after the Trump administration cut European access to Anthropic's newest models one day after the report came out.
Europe 2031 is the latest entry in a run of fictional AI-disaster scenarios written by little-known figures that have reached policymakers over the past year. The 2025 thought experiment AI 2027, which ends with a superintelligent system killing humanity, was read by US Vice-President JD Vance. Citrini Research's 2028 Global Intelligence Crisis, a story about white-collar unemployment, contributed to a stock dip in February.
Key Takeaways
- Europe 2031, a viral scenario published June 11 by Brussels-based researchers, puts the continent at 5% of global AI compute against 80% for the US.
- It spread through the G7 talks after Washington cut European access to Anthropic's Fable 5 and Mythos 5 models a day after the report's release.
- The authors urge Europe to build compute with American operators rather than fund a homegrown rival to Mistral.
- Skeptics counter that several US deals the scenario cites, including a $100 billion OpenAI-Nvidia pact, have already collapsed.
AI-generated summary, reviewed by an editor. More on our AI guidelines.
The main organization behind the report, the Arq Foundation, describes itself as neither an advocacy group nor a venture-backed startup and does not disclose who funds it. Daan Juijn, the foundation's research director, is one of the authors. Maximilian Negele, a contributor who left the US think tank Rand this year to work on the project, said he joined because of a translation barrier between Brussels and San Francisco, where the models are built, and that European inaction "just seemed like a slow-moving car crash to me."
The authors' recommendation breaks with the reflex that followed the Anthropic order. Rather than pour public money into Mistral, Europe's only frontier-model developer, they argue Europe should build far more AI computing capacity and use it to draw in American operators, giving the continent some bargaining power over their future decisions. "We've seen multiple well-known voices say things like we need to build our own European Mythos," Juijn told Euractiv, referring to Anthropic's Mythos model. "It's very easy to see where this idea is coming from, but it really underestimates how difficult it is." The build-out, the report says, would need public and private capital at a scale Europe has "not attempted in peacetime," along with deregulated zones for power and permitting like the data-center acceleration areas in the EU's proposed Cloud and AI Development Act.
Juijn also questioned the assumption that ASML, the Dutch company that builds the lithography machines used to make advanced chips, hands Europe a hold over the supply chain. "ASML is often called a choke point. I don't think it actually is a choke point," he said, arguing the United States would not feel the effect of any export restriction immediately. The report puts the funding distance in plain figures: OpenAI raised $122 billion in a single round in March, more than every European AI company has raised combined, while EU cloud providers' share of their home market fell from about 29 percent in 2017 to roughly 15 percent in 2022.
Get Implicator.ai in your inbox
Strategic AI news from San Francisco. No hype, no "AI will change everything" throat clearing. Just what moved, who won, and why it matters. Daily at 6am PST.
No spam. Unsubscribe anytime.
But several of the American projects the scenario holds up as proof of US dominance have since come apart. The $100 billion arrangement between OpenAI and Nvidia, last year's largest AI deal, collapsed in February. A $300 billion OpenAI-Oracle agreement looks doubtful, and OpenAI withdrew from the flagship Texas data-center project the scenario appears to reference. Negele said he would not rule out that "one or two AI companies might go bankrupt" but said the authors wanted to convey "a general feel for a version of what we think will happen."
Know someone who'd find this useful? ✉️ Email it to a friend in one click, or they can subscribe free here.
The document's prescription has critics. Pieter Garicano and Simon Grimm, writing on Silicon Continent, argued that protectionism and industrial policy will not close the gap, noting that Meta will spend more on chips this year, about $125 billion, than Germany spends on defense, around $114 billion, and is still trailing the leaders. At Gizmodo, Mike Pearl, who has published a book of hypothetical scenarios, wrote that the power such documents hold "is not good for the world," warning that speculative fiction is written backward from its ending and that readers mistake it for a forecast.
In Brussels the alarm has landed better than the specifics. Nicolás Casares, a Spanish member of the European Parliament who read the scenario, said some of its events "can happen" but that the authors "are increasing, a bit, the alarms in order to call our attention." The European Commission's Cloud and AI Development Act, which would create the accelerated build-out zones the authors want, still needs approval from all 27 member states and the Parliament, a process that usually runs 12 to 18 months.
Frequently Asked Questions
What is Europe 2031?
A speculative scenario published June 11 by AI researchers and investors at the Brussels-based Arq Foundation. It tells a fictional five-year story of Europe sliding into dependence on US and Chinese AI, paired with real data through mid-2026. It went viral during the G7 talks and was read by members of the European Parliament.
How much AI compute does Europe have?
The report puts Europe at roughly 5% of global AI computing capacity against about 80% for the United States. It also cites the gap between the largest US AI supercomputer, at 1,250 megawatts, and Europe's largest, at 83 megawatts.
What do the authors recommend?
Rather than fund a European rival to Anthropic's Mythos or pour money into Mistral, they argue Europe should expand AI computing capacity and use it to draw in American operators, building bargaining power. They call for capital at a scale Europe has not attempted in peacetime, plus deregulated zones for power and permitting.
Why did the scenario go viral now?
It was published one day before the Trump administration cut European access to Anthropic's Fable 5 and Mythos 5 models on June 12. The export order made the report's warning about dependence on US AI feel immediate, and it spread through the G7 talks in France.
What do critics say?
Some argue the remedy is wrong. Writers at Silicon Continent said protectionism will not close the gap, noting Meta will spend more on chips this year, about $125 billion, than Germany spends on defense, around $114 billion, and still trails. Others note several US deals the scenario cites have collapsed.
AI-generated summary, reviewed by an editor. More on our AI guidelines.
Marcus Schuler
Editor-in-Chief and founder of Implicator.ai. Former ARD correspondent and senior broadcast journalist with 10+ years covering tech. Writes daily briefings on policy and market developments. Based in San Francisco. E-mail: editor@implicator.ai
