EU settles on rules for generative AI, moves to surveillance
This article discusses the ongoing final negotiations in Brussels concerning the European Union's AI Act, a pioneering attempt at regulating artificial intelligence globally. Initiated in April 2021, these negotiations have recently made progress on controlling generative AI systems like ChatGPT but are still grappling with the contentious issue of AI for biometric surveillance, with a focus on national security and defence. There's a divide among EU member states, with countries like France and Germany, hosting leading AI firms, advocating for less restrictive regulations and self-regulation. Key concerns include the management of technological advancements, maintaining information on AI training models, and addressing the use of copyrighted materials. The talks, if inconclusive, might be postponed until after next year's European elections, potentially requiring a complete overhaul of the regulations due to rapid tech developments like Google's Gemini and ADM's new AI chip.