The European Union is preparing to open a Code of Practice in high-tech companies for general purpose AI (GPAI) as early as next week, ahead of the provisions of the major AI law that comes into effect on August 2nd.
Sources say official approval from EU member states and the European Commission could take place by July 22, and providers of AI systems such as CHATGPT and Gemini have settled down the way in which they voluntarily align their code. This document is intended to help you comply with the EU's groundbreaking AI law, which regulates AI based on risk levels.
Drafted by EU-appointed experts, the code sets voluntary rules and guidelines. Companies committing to it are expected to benefit from greater legal clarity and less testing, but companies that opt out may face more stringent surveillance.
Despite its spontaneous nature, the Code sparked controversy. Publishers, rights holders and tech companies have expressed concern about copyright rules and potential impacts on innovation. The initially anticipated release of the code in May was delayed due to these disputes.
Openai, developer of ChatGpt, has stated that he intends to sign the code in the end. “The code of practice opens the door for Europe to advance its EU AI Continental Action Plan,” Openai said in a statement.
Consumers and digital rights advocates are taking a cautious approach. Beuc's Cláudio Teixeira said the code was “a step in the right direction,” but warned that it should not dilute the binding protections of AI law. Laura Lazaro Cabrera of CDT Europe added that the code does not have any essential requirements to assess and mitigate risk in all cases.
As the EU prepares for the full implementation of AI Act, the world's first comprehensive AI law, future codes of practice are key tools for early compliance and serve as a preview of broader regulatory enforcement.
news.az
