EU members navigate the complexity of introducing AI regulations for platforms like ChatGPT and reach consensus.
As new developments in the field of artificial intelligence (AI) proceed at lightning speed, the European Union is also close to reaching a landmark agreement to regulate platforms like ChatGPT and other AI technologies.
EU lawmakers discuss regulating ChatGPT
Negotiators reached consensus on the regulation of generative artificial intelligence (AI) tools such as OpenAI Inc.’s ChatGPT and Google’s Bard during a lengthy meeting on Wednesday, December 6. The agreement, reached through discussions between the European Commission, the European Parliament and representatives of the 27 member states, brings them closer to finalizing the comprehensive legislation, the AI Act.
The negotiations, which began on Wednesday afternoon, highlight the complexity of the debate over artificial intelligence regulation. The proposed AI bill would be a key step in crafting a landmark AI policy that would set the tone for regulation of artificial intelligence tools around the world.
Furthermore, the discussion revealed the challenges policymakers face in finding a balance between protecting Europe’s AI startups and addressing potential societal risks. Efforts are currently underway to finalize the AI Act, reflecting the urgency of passing regulations before the upcoming European elections in June, which could bring about changes and hinder progress.
Interestingly, the discussion took place the same day that Google showed off new features for Gemini AI. Also, it happened just days after a major drama at OpenAI, where Sam Altman was ousted and then reinstated a few days later.
EU member states divided over AI regulations
The EU, like the United States and Britain, has grappled with the challenge of striking a balance between protecting AI startups such as France’s Mistral AI and Germany’s Aleph Alpha, and addressing potential societal risks.
The fight has become a key point of contention in the negotiations, with countries such as France and Germany voicing opposition to rules that could unfairly disadvantage local companies. Officials were growing optimistic about reaching a deal early Thursday, although the bill’s technical complexities still require further discussion in subsequent meetings.
EU policymakers have put forward a proposal outlining requirements for developers of AI models, including supporting tools such as ChatGPT. The plan requires maintaining model training information, summarizing the use of copyrighted material, and labeling AI-generated content. In addition, systems deemed to pose a "systemic risk" will need to cooperate with the Commission through an industry code of conduct to monitor and report any incidents caused by the models. #欧盟 #ChatGPT
