Restricting AI Training Rules Could Cost EU €600 Billion Annually, Study Warns
Tightening Europe's copyright rules for artificial intelligence training would slash €600 billion from the EU economy each year, according to an economic study released this week in Brussels. The findings come as the European Commission prepares to potentially rewrite the 2019 Copyright Directive, with a new legislative proposal expected in early 2027.
The study, conducted by Implement Consulting Group and the Ifo Institute, calculates the cost of restricting the EU's text-and-data-mining (TDM) framework - the copyright exception that currently allows companies to train AI models on large datasets across Europe.
Market Uncertainty Threatens AI Development
The Commission's call for evidence on reopening copyright rules has created immediate uncertainty. This timing is particularly fraught: the EU AI Act's copyright provisions take effect in August, and policymakers are now signaling potential changes before those rules even begin operating.
The study estimates that generative AI could generate €1.65 trillion in annual value for the EU. Restricting the TDM framework would put more than a third of that opportunity at risk.
Limiting data access would degrade AI model capabilities by up to 50%, the research found. This directly threatens €275 billion in annual value across healthcare, finance, legal services, and other professional sectors that rely on complex reasoning.
Four Specific Economic Risks
A single year of regulatory delay would cost the EU €175 billion in lost AI adoption. Currently, 53% of EU firms cite unclear legal consequences as a primary barrier to deploying AI technologies.
Restricting data access would drive €60 billion annually in local AI development away from Europe. Top-tier talent, developers, and venture capital would migrate to jurisdictions with more permissive rules.
Weaker AI models would stall innovation in knowledge-intensive industries. Pharmaceuticals and advanced manufacturing alone would lose €90 billion per year in research efficiency.
Three Policy Recommendations
The study recommends that policymakers maintain the commercial TDM exception to give AI developers legal certainty. Companies need confidence that they can scale operations without legal jeopardy.
Second, the report urges the EU to keep opt-out mechanisms workable by endorsing globally recognized machine-readable standards like robots.txt, rather than unproven alternatives that could create compliance nightmares.
Third, policymakers should avoid rigid mandatory licensing frameworks that would create transactional bottlenecks. Voluntary commercial data-sharing partnerships should continue developing naturally.
Industry Perspective
Boniface de Champris, AI Policy Lead for the Computer & Communications Industry Association Europe, said the Commission should not reopen the rulebook as the AI Act's copyright provisions are about to take effect. "Doing so would send the wrong signal to the market and suggest the EU does not trust its own rules," he said.
Martin H. Thelle, Senior Partner at Implement Consulting Group, added that Europe has a large AI opportunity, but it will not materialize if policymakers make training and deploying competitive AI systems harder. "The region risks slower adoption, weaker capabilities, and less innovation," he said.
For professionals in government and finance evaluating AI policy, understanding these economic trade-offs is essential. The decision Brussels makes in 2027 will determine whether Europe remains competitive in AI development or cedes ground to jurisdictions with clearer rules.
AI for Executives & Strategy courses can help decision-makers understand the policy implications and competitive dynamics shaping AI regulation across regions.
Your membership also unlocks:
