Last week, Meta declared its decision not to endorse the European Union’s (EU) Code of Practice for general-purpose artificial intelligence (GPAI) models. Earlier this month, the European Commission announced that it received the final version of the GPAI Code of Practice, a voluntary framework designed to assist the industry in adhering to the AI Act’s regulations, which will take effect on August 2. The tech company headquartered in Menlo Park cited “a number of legal uncertainties for model developers” as the reason for its refusal to sign.
Joel Kaplan, Meta’s Chief Global Affairs Officer, revealed the company’s stance in a LinkedIn post. He indicated that the decision followed a thorough examination of the document, stating, “Europe is heading down the wrong path on AI.”
The GPAI Code of Practice comprises three main chapters: Transparency, Copyright, and Safety and Security. It outlines guidelines for creating a user-friendly model documentation form, compliance with EU copyright law, and ensuring that large language models (LLMs) do not pose systemic risks to fundamental rights and safety. The Code of Practice serves as a preliminary framework for companies to prepare for the AI Act, and it carries no legal enforcement.
Kaplan expressed criticism of the Code of Practice, arguing that it exceeds the intended scope of the AI Act. Notably, several major companies in Europe, including Airbus, Lufthansa, Mercedes-Benz, and Siemens Energy have signed an open letter calling on the European Commission to “stop the clock” on the AI Act.
In his post, Kaplan conveyed that the company aligns with the concerns articulated by these firms. He warned that the EU’s “over-reach” could hinder the innovation and deployment of advanced AI models within Europe. Interestingly, OpenAI has already expressed its commitment to signing the Code of Practice.
Despite the objections from various stakeholders, reports indicate that the European Commission intends to proceed with the AI Act according to the previously established timeline. Commission spokesperson Thomas Regnier emphasized, “I’ve seen, indeed, a lot of reporting, a lot of letters and a lot of things being said on the AI Act. Let me be as clear as possible; there is no stop the clock. There is no grace period. There is no pause,” he is quoted by Reuters.
