Yesterday, every major newspaper in the UK had the exact same headline, demanding the government “Make It Fair” by ensuring that the country’s copyright laws make it impossible, or at least very expensive, for AI models to be trained there.

This stunt is apparently a response to the UK’s open consultation on the topic of AI and Copyright that welcomed anyone with an interest in the subject to weigh in on the government’s proposed “plan to deliver a copyright and AI framework that rewards human creativity, incentivises innovation and provides the legal certainty required for long-term growth in both sectors.”

There are, of course, legitimate concerns surrounding AI development and its impact across all sectors of society. While this consultation is ostensibly focused on commercial players, libraries and archives—and the public they serve—would be affected by possible sweeping new copyright rules. These are mission-driven organizations that can help with giving citizens historical background to current issues and help counter propaganda campaigns, among other social benefits.

Without a change to its current law, universities and libraries in the UK might not be able to work with universities in other countries on many research projects. While we urge the UK government to retain some form of noncommercial exception for text and data mining (TDM), as we point out in our submission, the current version is burdensome—for example, requiring that any digitized versions of physical materials made for the purpose of one project be scanned again and again for any other projects. Non-commercial actors, many of which are government-funded, should be encouraged to work together and be efficient, not to have to digitize the same materials over and over.

Unfortunately, libraries and other cultural heritage institutions do not control the front pages of newspapers. Nevertheless, we hope the UK government will pay attention to what such noncommercial public interest organizations have to say about the future of our information ecosystem and the development of AI tools that work for everybody.

You can read the Internet Archive’s full comments here.