In this new episode of the privacyespresso series, we share insights on the AI Act and the latest legislative updates. Join Siyanna Lilova, from Timelex and Matthew Newman, global chief correspondent for MLex, as they delve into the EU’s Proposed Artificial Intelligence Act.
Proposed by the European Commission in 2021, the AI Act is a landmark in the global regulation of Artificial Intelligence. The AI Act’s general structure is a risk-based approach to regulation, from banning AI systems with unacceptable risks, to high-risk AI systems subject to a wide range of obligations for providers and users. Companies will be required to get AI systems certified by national authorities before putting them on the market for use in “high-risk” scenarios, such as insurance assessments, school admissions, hiring and firing or the judicial system.
The AI Act has entered the final phase of negotiations between the European Commission, the European Parliament and EU governments. The exact content of the high-risk AI list is not yet certain. Talks will also focus on which AI systems should be banned because of unacceptable risks, such as distorting human behaviour by subliminal techniques, social scoring and real-time biometric identification.
Some of the main sticking points are the definition of AI, the risk classification of AI, and how generative AI and facial recognition technology in public spaces should be regulated. Negotiators are also debating how to regulate “foundation models” that form the basis of Generative AI systems, such as ChatGPT, a #chatbot that uses natural language processing to create humanlike chats. ChatGPT, which was released last November, quickly attracted millions of users, stoking concerns about the power of generative AI. The most difficult disagreement is over parliament’s push to ban police use of real-time facial recognition systems in live #cctv feeds. EU governments are seeking an exception for the use of AI for real-time biometric identification for law enforcement. The goal is to reach a deal by year’s end. A deal needs to be done by February 2024 before European Parliament members begin campaigning for elections in June 2024.
Don’t miss our upcoming PrivacyRules conference for a deeper dive into the AI Act and international AI regulation!
If you like this quick session, join us and Matthew Newman in person at the upcoming PrivacyRules conference in Brussels on November 14. We will host sessions with top-level experts and have a networking cocktail together.
You can find more details and register here 👉 https://lnkd.in/dsgXv-Wb