United KingdomText and Data Mining for Non-Commercial Research
Copyright, Designs and Patents Act 1988, c. 48, s. 29A
In force since .
A copyright and text and data mining (TDM) rule binding public and private bodies.
- Audit expectation
- none
What it requires
- Obtain a licence or the rights holder's consent before crawling third-party websites to collect data for commercial AI model training, since the non-commercial-research text-and-data-mining exception does not cover commercial use.
- Do not transfer a copy made under the non-commercial-research exception to anyone else, or use it for any purpose beyond that research, without the copyright owner's authorisation.
These lines are LexLint's reading of the law's text.
In the frameworks
- NIST AI RMF
- MAP 4.1Approaches for mapping AI technology and legal risks of its components – including the use of third-party data or software – are in place, followed, and documented, as are risks of infringement of a third party’s intellectual property or other rights.
Our reading of where this law's duties sit in each framework. It never says that running a control meets the law. How we read a framework against the law
What it makes you log
Logging duty
Section 29A permits a copy for computational analysis for non-commercial research on conditions: lawful access, a sole non-commercial research purpose, and a sufficient acknowledgement on the copy. Subsection (2) limits transfer and further use of the copy and subsection (5) makes contract terms against the exception unenforceable. The acknowledgement attaches to the copy and is not a log of what was copied or when, and no clause requires a record to be kept or produced. The section has no territorial clause of its own.
- Kind
- None
- As of
What this law does
Section 29A permits a person with lawful access to a work to make a copy of it for computational analysis, without infringing copyright, but only where the copy is made so that a person with lawful access can carry out that analysis for the sole purpose of non-commercial research, and the copy carries a sufficient acknowledgement.
The copy becomes an infringing copy if it is transferred to anyone else, other than with the copyright owner's authorisation, or used for any purpose beyond that non-commercial research purpose. A contract term purporting to prevent or restrict the making of a copy that this section would otherwise permit is unenforceable.
Because the exception is limited to non-commercial research, crawling third-party websites to collect training data for a commercial AI model does not fall within it, and depends instead on a licence or the rights holder's consent under the general copyright framework.
The government's Report on Copyright and Artificial Intelligence, published following its 2024-25 consultation, stated that a broad text-and-data-mining exception with a machine-readable opt-out is no longer the government's preferred way forward, so the position described here is expected to remain the status quo for the near term.
When LexLint raises it
When your app profile says your app crawls the web or trains models.
Back to the example · Lint your app
Page updated .