Europe confident its rules can thwart rogue AI (www.rnz.co.nz)

The European Union (EU) is expressing confidence in its comprehensive AI regulations, as outlined in the AI Act, to manage the risks posed by rogue AI systems. Henna Virkkunen, the EU tech chief, highlighted that the Act, which governs AI based on risk levels, is well-equipped to address concerns that have arisen following incidents involving major players like OpenAI and Anthropic. The legislation covers the entire lifecycle of AI models and emphasizes safety and security, responding to the growing global anxiety about AI management. Virkkunen also mentioned that the EU has engaged with over 30 AI companies, including those from the U.S. and China, to investigate their safety protocols and compliance measures. This outreach could lead to significant regulatory actions, including hefty fines for negligence. Dismissing claims that the EU’s rules are outdated, she asserted that the legislation is adaptable to evolving risks posed by advanced AI technologies. By utilizing a scientific panel of 60 experts and ensuring ongoing monitoring of AI models, the EU aims to remain at the forefront of responsible AI governance, addressing both domestic and global concerns.