Is OpenAI’s promised copyright protection stalled for good?

According to an exclusive TechCrunch article, OpenAI has not delivered the promised ‘Media Manager’ opt-out tool, which was teased in May 2024 to enable creators and content owners to prevent their works from being used in AI training. As 2025 begins, insiders indicate that this feature is not a priority for the company.

OpenAI delays media manager tool for copyright protection

The ‘Media Manager’ tool was designed to help identify copyrighted content, including text, images, audio, and video, and was intended to allow creators to set preferences regarding the use of their works. The announcement was made in response to extensive criticism over OpenAI’s utilization of copyrighted material for training its models, like Sora, which generates content trained on billions of webpages and may produce near duplicates of copyrighted works.

Despite the initial enthusiasm surrounding Media Manager, reports suggest it has not been a significant focus internally. A former OpenAI employee told TechCrunch, “I don’t think it was a priority. To be honest, I don’t remember anyone working on it.” A contact closely coordinating with OpenAI further indicated that there have been no recent updates regarding the tool. Fred von Lohmann, a member of OpenAI’s legal team who was involved in Media Manager, transitioned to a part-time consultant role in October.

Italy slaps OpenAI with a €15M fine over GDPR breach in ChatGPT

Since the announcement, OpenAI has not made any public statements regarding the progress of Media Manager. The company missed a self-imposed deadline to have the tool functioning “by 2025,” leading to uncertainty about its future. OpenAI has implemented some ad hoc measures for creators to opt out of AI training but these have been criticized for being inadequate. A submission form for artists to flag their work for exclusion has been perceived as cumbersome, and webmasters can block data scraping but lack specific mechanisms for different media types.

Experts question whether Media Manager can effectively resolve the legal issues related to intellectual property usage. Adrian Cyhan, an IP attorney, expressed skepticism about OpenAI’s ability to navigate the complexities of creator protections. Ed Newton-Rex, founder of Fairly Trained, noted that the burden of control may unfairly shift to creators who might not be informed about the tool’s existence.

OpenAI is currently involved in class-action lawsuits from various creators, including authors and media companies, who allege that their works were used without permission. The company maintains that its AI-generated outputs qualify as fair use by producing transformative works. OpenAI asserts that developing competitive AI models without leveraging copyrighted materials would be infeasible and emphasizes the need for a thorough approach to data sourcing and compliance with legal requirements.

Featured image credit: OpenAI