Industry Leaders Call for Clearer AI Regulations in Europe

In a public appeal, Meta and more than 50 companies, researchers and industry leaders, including Ericsson, Spotify and Harvard University, issued an open letter urging the European Union to address its fragmented approach to AI regulations. The letter, published in the Financial Times, highlights growing concerns that inconsistent and unclear regulations are putting Europe at risk of falling behind in the global AI race.

The Call for Regulatory Certainty

The letter argues that Europe’s slow and unpredictable regulatory decision-making, especially from European Data Protection Authorities (DPAs), is creating uncertainty around what data can be used to train AI models. This, according to the signatories, jeopardises Europe’s ability to fully harness AI’s potential to boost productivity, fuel economic growth and advance scientific research.

Without clearer guidance and harmonised decisions, companies fear they will not be able to effectively utilise European data for AI development. This could lead to a significant gap between Europe and regions like the U.S., China and India, where AI technologies are advancing rapidly. The letter stresses the need for “harmonised, consistent, quick, and clear decisions” to ensure that European data can be used responsibly and efficiently to train AI models for the benefit of businesses and citizens alike.

The Importance of Open and Multimodal AI Models

The letter focuses on two key technological advancements that Europe risks losing out on: open models and multimodal models.

Open models allow organisations to freely modify and use AI frameworks, empowering them to control data internally without the need for external dependencies. These models not only enhance data sovereignty but also enable widespread innovation across different sectors.

Multimodal models, which can process text, images and speech, represent the next big leap in AI development. These models offer a significant upgrade from text-only models and have the potential to transform industries by improving productivity, facilitating research and adding economic value across sectors.

The letter warns that Europe is at a crossroads. If it fails to establish a clear regulatory framework for AI, it could miss out on the numerous benefits offered by these advanced models. In contrast, regions like the U.S. and China are already embracing these technologies, positioning themselves as leaders in the AI revolution.

The Role of GDPR and the AI Act

A key issue raised in the letter is the impact of the 2018 General Data Protection Regulation (GDPR) on AI development. While the GDPR has been vital in protecting data privacy, recent decisions by DPAs have introduced uncertainty around the use of personal data in AI training. The signatories call for a modern interpretation of GDPR provisions that strikes a balance between data protection and AI innovation, ensuring that European AI models remain competitive globally.

Moreover, the EU’s AI Act, aimed at preventing the misuse of AI technology, has added another layer of complexity. While the act is crucial for regulating AI responsibly, companies are concerned that the current regulatory environment is stifling innovation. This has already led to delays in the European release of major tech products. Meta, for example, postponed the launch of its Twitter alternative, Threads, while Google delayed the introduction of AI tools in the region due to regulatory uncertainties.