OpenAI is rolling out an invisible watermark system for ChatGPT-generated text in the European Union to comply with the AI Act regulations. The company has also developed a detection tool capable of identifying watermarked content. However, OpenAI’s own testing shows that simple synonym swaps can effectively bypass this system. This deployment addresses the transparency requirements imposed by the European AI Act, which obliges generative AI developers to indicate the synthetic origin of their content.
Source: Read the original article
Disclaimer: this content is for information purposes only and is not financial advice. Cryptocurrencies are highly volatile: you may lose all of your capital. Always do your own research. Legal notice
