The European Commission has opened two stakeholder consultations under the EU Artificial Intelligence Act, targeting both copyright-related obligations for General Purpose Artificial Intelligence providers and the detailed design of Artificial Intelligence regulatory sandboxes. The first consultation, which closes on 9 January 2026, focuses on how providers of general-purpose Artificial Intelligence models should implement technical protocols to respect rightsholders’ reservations of rights against text and data mining under the Artificial Intelligence Act and the associated General Purpose Artificial Intelligence Code of Practice. The second consultation, which closes on 6 January 2026, seeks feedback on a draft implementing act that sets common rules for establishing and operating Artificial Intelligence regulatory sandboxes across the European Union.
Launched on 1 December 2025, the copyright consultation is intended to support the Artificial Intelligence Act’s requirement that providers of general-purpose Artificial Intelligence models adopt a policy to comply with EU law, including identifying and honoring reservations of rights expressed pursuant to Article 4(3) of Directive (EU) 2019/790. In line with the General Purpose Artificial Intelligence Code of Practice, signatories commit to use appropriate machine-readable protocols for signaling such reservations, including robots.txt and subsequent IETF standards. The Commission invites rightsholders, model providers, civil society organizations, standardization bodies, and other stakeholders to comment on the technical feasibility and likely uptake of different reservation-of-rights solutions identified in the EUIPO study on “Development of Generative Artificial Intelligence from a Copyright Perspective,” with the goal of finding standardised, interoperable mechanisms that work across media, languages, and sectors.
Stakeholders have until the 9 January 2026 to respond to the copyright consultation and may also express interest in Commission-organized workshops, which signatories to the General Purpose Artificial Intelligence Code of Practice will be invited to by default. After the consultation, the Commission will publish a list of generally agreed machine-readable reservation-of-rights solutions, which will be reviewed regularly at least every two years in line with updates to the Code. In parallel, a consultation launched on 2 December 2025 covers a draft implementing act for Artificial Intelligence regulatory sandboxes that defines their establishment, development, implementation, operation, and supervision, including participation terms and selection criteria. These sandboxes are intended to provide a controlled environment for businesses to test innovative Artificial Intelligence systems under regulatory oversight, support compliance with Article 57, which requires Member States to establish at least one national Artificial Intelligence regulatory sandbox by 2 August, 2026, and advance the Artificial Intelligence Act’s broader objective of fostering innovation while ensuring adherence to its rules. Stakeholders have until 6 January 2026, to comment on the sandbox framework, including participation criteria, scope of permitted activities, and post-participation obligations.
