Currently, when discussing AI content, a very big question is who generated this content? What data was used? Has it been contaminated? Especially as deepfakes, forged images, and model theft become increasingly easy, "content ownership" has become a problem that must be solved. The PoA concept proposed by OpenLedger this time is specifically developed to solve this problem. PoA (Proof of Attribution) is a type of authentication for "AI traceability". Whether it is text, images, audio, or video, any content generated by AI can be automatically accompanied by a PoA (Proof of Attribution) through PoA technology Who generated the watermark+how it was generated+generation time This watermark cannot be directly seen, but it can be verified in digital form. In human terms, when you see any object that appears to be generated by AI, whether it is an article, image, or audio or video, this technology can be used to check whether it is generated by AI, which large model was used to generate it, and even accurately determine which version of the model it is, such as GPT 3.0 version. This application has a great ability to distinguish between the current proliferation of AI, such as articles on Twitter or borderline photos of young ladies, which can clearly know whether they are real or generated by AI. The action of adding a 'watermark' is the technique of Infini gram. This technology is essentially a content level fingerprint system that can embed a unique identifier without affecting the quality of images or text, and this identifier is resistant to compression, editing, etc. and will not be cleared. It's like a signature at the code level. The use of PoA+Infini gram technology is like issuing a "digital birth certificate" to every AI generated content. Under this mechanism, the output of the AI model is no longer untraceable, but every content generated by AI can leave an identifier, making it easy to determine whether it was generated by AI. Especially for content creators, it increases their competitiveness in the market. So personally, I also hope that the approach of @ Openledger HQ PoA can be applied to more large models, allowing more people to easily identify AI products.