OpenAI teams up with its rival Google: ChatGPT gains full access to DeepMind’s invisible watermark and launches an image authenticity verification tool.

AIMPACT News. On May 20 (UTC+8), according to Beating monitoring, OpenAI announced a partnership with its former rival, Google DeepMind. It fully integrated SynthID invisible watermarks into images generated by ChatGPT, Codex, and via the API, and officially became a C2PA image credential standard-compliant product, while also rolling out a public preview of an image authenticity verification tool.

As an industry-standard C2PA metadata relies on cryptographic signatures to record provenance, this metadata is highly likely to be lost when users download images, take screenshots, or convert image formats. To address this loophole, OpenAI introduced SynthID invisible watermark technology developed by Google DeepMind, embedding identifiers that are invisible to the naked eye directly into image pixels. Even after secondary processing such as screenshots, SynthID can still be preserved. A two-layer defense mechanism combines the rich context of metadata with the persistence of invisible watermarks.

Along with this update, OpenAI also launched a public preview of an official website to verify authenticity (openai.com/research/verify). Users can upload images directly on the web page, and the tool will detect whether the image contains OpenAI’s Content Credentials credentials or SynthID invisible watermarks. However, OpenAI is very cautious in its detection conclusions. Because watermarks and metadata may still be forcibly stripped in extreme cases, even if the tool does not find any anti-counterfeiting signals, it cannot conclusively determine that the image was definitely not generated by AI.

(Source: BlockBeats)