Recently, eight U.S. newspapers sued OpenAI for IP infringement related to the use of generative AI. OpenAI and other generative AI vendors argue that fair use shields their practice of scraping public data for training without compensating or crediting artists. However, to appease critics, OpenAI has allowed artists to opt out of and remove their work from the data sets used to train its models. Despite these efforts, some content creators argue that these tools are not sufficient and do not address scenarios where creators' works are quoted, remixed, or reposted on platforms they do not control.
Key takeaways:
- OpenAI is developing a tool called Media Manager to allow creators and content owners to control how their works are used in generative AI research and training.
- The tool is expected to be in place by 2025 and will identify copyrighted text, images, audio and video across multiple sources, reflecting creator preferences.
- OpenAI has faced criticism and legal action for its practice of scraping public data for AI training without compensating or crediting artists, arguing that it falls under fair use.
- Existing tools allowing artists to opt out of data sets and website owners to control scraping have been deemed insufficient by some content creators, and OpenAI acknowledges that its solutions do not cover all scenarios.