US Tech Giants Push for Simple AI Code of Practice

The European Commission is finalizing a voluntary Code of Practice on General-Purpose AI (GPAI) to support compliance with the EU’s AI Act. Recent meeting minutes reveal that major US technology companies—including Amazon, IBM, Google, Meta, Microsoft, and OpenAI—have urged the Commission to keep the code straightforward. Their primary concerns center on minimizing redundant reporting and avoiding excessive administrative requirements that could hinder innovation and create unnecessary burdens.

The Code of Practice is intended to provide guidance for AI model providers, such as those developing large language models like ChatGPT, in meeting the requirements of the AI Act. The final draft was delayed after the Commission extended consultations in response to numerous stakeholder requests. Over 1,000 participants, including experts and industry representatives, have contributed feedback through plenary sessions and workshops since the process began last September.

US tech firms have emphasized that the final text should grant signatories adequate time to implement its commitments and should remain within the intended scope of the AI Act. They caution that overreaching requirements could stifle innovation and place disproportionate demands on providers. Meanwhile, publishers and creative industry representatives, such as CISAC president Björn Ulvaeus, have voiced concerns that Big Tech-driven proposals may undermine copyright protections under the new regime.

The European Commission aims to publish the revised Code of Practice before the summer, with the rules on GPAI tools set to take effect on 2 August. The AI Act itself, which categorizes and regulates AI systems based on societal risk, entered into force in August of the previous year. Its provisions will be phased in, with full applicability expected by 2027.