OpenAI has fallen short on its promise to deliver Media Manager, a tool intended to help photographers exclude their work from AI training data, missing its self-imposed 2025 deadline.
The company announced Media Manager in May as a solution to identify copyrighted content across text, images, audio, and video formats. The system was positioned as OpenAI's response to mounting criticism and copyright disputes surrounding its AI training practices.
However, development appears to have stalled, with a former OpenAI employee revealing to TechCrunch that they "don't remember anyone working on it," suggesting the project was not a priority for the company.
When OpenAI launched DALL-E 3, its latest image generation model, the company offered photographers an opt-out option. This process required creators to submit individual works with detailed descriptions – a tedious approach that many found impractical.
Ed Newton-Rex, founder of Fairly Trade, criticized this approach, noting that "most creators will never even hear about it, let alone use it," while the system would still be used to "defend the mass exploitation of creative work against creators' wishes."
The controversy highlights ongoing tensions between AI companies and content creators, who object to their work being used for AI training without consent or compensation. Media Manager was supposed to address these concerns by providing a streamlined opt-out process.
IP attorney Andrian Cyhan from Stubbs Alderton & Markiles points out the complexity of such a system, comparing it to content ID challenges faced by major platforms like YouTube and TikTok. He notes that compliance across different jurisdictions adds another layer of difficulty.
OpenAI's last update on Media Manager came in August, when a spokesperson stated the tool was "still in development." Since then, the company has remained silent on the project's status, leaving photographers and other content creators without the promised solution.