Consensus Eludes UK Government, AI Companies, and Creative Organizations on AI Copyright Code

After nearly a year of deliberations, the Intellectual Property Office (IPO) of the UK has failed to broker a consensus between government authorities, AI industry leaders, and creative organizations regarding a proposed code of practice for training AI models on copyrighted material. 

The aim of these discussions was to establish clear guidelines for the practice of text and data mining, particularly concerning AI models trained on copyrighted books, images, and films. 

Regrettably, the consortium mediated by IPO has not agreed on a voluntary code of conduct. Consequently, the responsibility has been returned to the Department for Science Innovation and Technology officials, with no imminent prospects of definitive policies being set.

A concern for creative professionals

This impasse in negotiations is a significant setback for creative professionals who are increasingly concerned about the potential misuse of their work. AI tools, such as OpenAI’s ChatGPT and Stability AI’s text-to-image generator Stable Diffusion, often rely on data scraped from the web, leading to generating content that may closely resemble the source material. 

This practice has sparked numerous concerns among creators who fear their work could be copied, reproduced, and distributed without proper attribution or compensation.

In 2023 alone, the tech industry witnessed a surge in copyright litigation, with allegations of tech firms appropriating artists’ work to train their AI models. One of the most high-profile cases unfolded in the United States, where The New York Times took legal action against OpenAI and Microsoft for copyright infringement. These lawsuits underscore the pressing need for clear regulations surrounding AI and intellectual property rights.

AI’s rapid expansion and legal lag

The entertainment industry has seen rapid integration of AI technologies in recent years, from automated audiobooks and voice assistants to deepfake videos and text-to-speech tools. Unfortunately, legal frameworks have struggled to keep pace with these technological advancements. The absence of well-defined guidelines and regulations has left both creative professionals and AI developers in a state of uncertainty.

Recognizing the urgent need for legal reform, Equity, a trade union representing 50,000 performers and creative practitioners in the UK, has initiated the “Stop AI Stealing the Show” campaign. The campaign aims to lobby the government for comprehensive updates to the law to safeguard the livelihoods of artists and creators.

 Equity has expressed its readiness for “industrial action,” reminiscent of the 2023 Hollywood strikes, should key agreements not be reached regarding AI and intellectual property. Liam Budd, an official at the trade union, has criticized the government’s cautious approach to AI regulation, advocating for more proactive measures.

Beyond the creative industry, pioneers in generative AI have also voiced their concerns. Ed Newton-Rex, a notable figure in the AI field, left Stability AI in November due to the startup’s use of copyrighted content. 

In January, he launched a non-profit organization called “Fairly Trained,” which aims to certify AI companies that ethically source their data. Newton-Rex’s move underscores the growing momentum behind the push for ethical AI practices and copyright protection.