AI companies are now working on new ways to identify content created by their AI models. After Anthropic introduced a new system to identify text generated by Claude, OpenAI is also expanding its tech to track the origin of AI-generated content. Sam Altman’s OpenAI already uses C2PA Content Credentials and SynthID to add provenance signals to AI-generated images. These signals can help identify whether an image was created using OpenAI tools such as ChatGPT, Codex or the OpenAI API. The ChatGPT maker has now expanded this system beyond images. In an updated policy page, the AI startup said supported audio generated using its tools, including ChatGPT and the OpenAI API, now also includes SynthID watermarks. OpenAI has also launched a verification system that can check supported images and audio for these watermarks. Is there need why the models keeps on evolving ?
