There are divisions in European Union law about how to regulate AI applications and this has caused problems for lawmakers trying to make a deal on draft legislation. They are arguing over how the law should approach upstream AI model makers. French startup Mistral AI is at the center of this debate because they have been lobbying against a European Parliament proposal on regulating generative AI. Specifically, they have issues with the parliament’s push for transparency requirements for foundational model makers, as well as their suggestion to regulate relationships in the AI value chain. The dispute is complicating efforts to reach a political compromise – a compromise that these EU lawmakers have to reach soon. The Commission originally proposed the framework for regulating AI back in April 2021, but much has happened in the realm of large language models (LLM) and generative AI since then. The European Parliament wants to make sure the AI Act keeps up with these fast-moving developments, but their efforts have been met with opposition from some Member States. Mistral has concerns that the proposed legislation will give unfair advantages to US AI giants, create a convoluted bureaucracy, and mean a loss of healthy competition for homegrown AI startups. The startup also argues that deploying AI app makers should be responsible for enforcing hard laws, not upstream model makers. This position is backed up by Cédric O, formerly a digital minister in the French government and now co-founder at Mistral. He also believes that lawmakers are misunderstanding the technology by thinking that safety and trustworthiness can be achieved by imposing obligations upstream. US AI giants are also hesitant about the prospect of tighter regulations in Europe, with OpenAI’s Sam Altman even suggesting they could leave the region if the AI rules aren’t to their liking. France and Germany are reportedly leading the push-back against tighter regulation of foundational models, while other countries like Italy, the Netherlands and Denmark are also opposing it. Mistral feels that it is crucial to find a balance between how the technology is developed in Europe and how consumers and citizens are protected.
Related Posts
Threads introduces feature allowing users to disable cross-posting to Facebook and Instagram
In response to user feedback, Instagram’s Threads will now let users turn off automatic sharing of their Threads posts to…
Oxford Study Finds No Harmful Effects of Internet Use on Mental Health, But Highlights Research Limitations
The article is too long to be rewritten in a single attempt. The original article contains 1000+ words, which exceeds…
Kapital Raises $165 Million to Empower Latin American SMBs with Financial Clarity
Bank and technology platform Kapital continues to rake in venture capital, grabbing another $40 million in Series B dollars and…