Following a three-day marathon of negotiations, the EU AI Act finally passed over the weekend with Carme Artigas, the Spanish secretary of state for digitalization and artificial intelligence, calling it “a historical achievement, and a huge milestone towards the future!”
Among other agreements, the EU AI Act requires systems labeled as ‘high-impact general purpose AI models’ to adhere to transparency standards, while those classified as ‘high risk’ are subject to extra requirements including risk management, monitoring of severe incidents, evaluating the AI model, and implementing red teaming strategies. In other words, the EU AI Act was a major development in the field of AI regulation.
The EU AI Act, while initially celebrated as a monumental event in the AI community, lost some media momentum when it became clear that it is a provisional agreement. This means that it could take months before the bill receives its final wording and even longer before it is enforced. The Act won’t take effect until two years after final approval from European lawmakers. This timeline may seem extensive, considering the rapid pace of advancements in AI technology.
Mistral: Shaking Up the AI Landscape
The EU AI Act was quickly overshadowed by the actions of Mistral, a Paris-based open source model startup. Mistral made waves by releasing a new LLM on Friday, accompanied only by a torrent link. This move was followed by the announcement of a $415 million fund-raise, valuing the startup at about $2 billion.
Mistral had been actively involved in the negotiations surrounding the EU AI Act, lobbying against certain regulatory proposals. The fact that Mistral had deep government connections, with Cedric O, France’s former state secretary for digital, leading its lobbying efforts, raised eyebrows among industry observers.
Mistral’s lobbying efforts seem to have paid off, as the agreed-upon EU AI Act provides exemptions for open source models (excluding those deemed to pose a systemic risk). Despite Mistral’s successes, it is clear that the EU AI Act is only the beginning of AI regulation in the European Union.
A Global Race for AI Sovereignty
The EU’s desire to avoid being left behind in AI development while simultaneously imposing regulations reflects the ongoing debate surrounding AI sovereignty. With Germany’s Vice Chancellor and Minister for Economic Affairs Robert Habeck emphasizing the importance of European companies taking the lead in AI, it is evident that achieving AI sovereignty is a key priority.
AI regulation, whether in the EU, the US, or elsewhere, is playing catch-up and will continue to do so. As AI researchers gather at conferences like NeurIPS to showcase their advancements, policymakers are grappling with the challenge of balancing innovation with regulation. The need for speed in implementing effective regulations is highlighted by Mistral’s rapid rise.
As the AI landscape evolves, it is clear that the EU AI Act and Mistral’s achievements mark significant milestones in shaping the future of AI regulation and development.
