Skip to main content

Written question asked by Lord Clement-Jones (Liberal Democrat) on Monday, 20 April 2026, in the House of Lords. It was due for an answer on Tuesday, 5 May 2026. It was answered by Baroness Lloyd of Effra (Labour) on Monday, 27 April 2026 on behalf of the Department for Science, Innovation and Technology.


Artificial Intelligence: Copyright

Question

To ask His Majesty's Government what binding contractual terms, conditions of use, monitoring arrangements or audit mechanisms apply to access to the AI Research Resource supercomputer network, to ensure that public compute is not used for the training or development of AI models in a manner that would infringe UK copyright law; and what consequences apply in the event of non-compliance.

Answer

Access to the AI Research Resource (AIRR) is governed by binding contractual and operational conditions to ensure that publicly funded compute is used responsibly and in line with UK law, including UK copyright law.

All users are required to enter into formal access agreements with the relevant host institutions and funders. The AIRR is currently delivered through systems hosted by the University of Bristol (Isambard‑AI) and the University of Cambridge (Dawn), with funding provided by UK Research and Innovation (UKRI) on behalf of the Government. DSIT have been clear that copyright rules should also be respected: use of copyright works to train AI in the UK will require a licence unless an exception applies.

Where there is evidence or reasonable suspicion of breach of the AIRR conditions of use, remedial actions may include investigation by the host institution, suspension or termination of access, withdrawal of compute allocations, and, where appropriate, the application of contractual remedies.


Secondary information

Type
Written question
Reference
HL16528
Session
2024-26
Subjects
Computers Copyright Research Artificial intelligence
Link
View this Written question on www.parliament.uk