The New York Times rejects OpenAI and Microsoft’s argument that the unlicensed use of copyright-protected content is “fair use” (US definition) because it is being used for a “transformative purpose”.

The New York Times does not believe their content should be used in this way, claiming that the scraping of its material to train large language models has resulted in AI users being able to access their trusted content without a paywall.

Whilst some companies have restricted employees from using ChatGPT due to fears of feeding the systems confidential data and some, including Bloomberg, Reuters, the BBC, and the Guardian, have completely blocked the tech that allows AI to use their content for training purposes.

Meanwhile, in an evidence submission to a parliamentary inquiry, OpenAI said it would be “impossible to create AI tools like ChatGPT without copyrighted material”. The PPA also submitted evidence to this inquiry and will monitor the publication of its findings.

At present, there is no AI regulation in the UK and the current government has said there’s no “rush to regulate”. The PPA advocates for, and has been asserting to the government, the need for AI regulation, particularly in relation to copyright and digital competition. We will continue to advocate for this with all political parties as we approach the 2024 election.

The PPA will be lobbying parliamentarians amidst current litigation overseas, doing so by addressing “fair dealing” (the UK equivalent to “fair use”) and calling for regulatory mechanisms to support all publishers who may be affected by data scraping.

To read The New York Times lawsuit in full, click here.

If you wish to discuss these issues further with our public affairs team, please contact eilidh.wilson@ppa.co.uk or natasha.anku@ppa.co.uk