While Reddit enjoys making fun of AI news, there comes a major twist for content generated that way. In fact, seven major companies in the sector say they are ready to introduce gods watermark to make AI content “recognizable”.
In this regard, as also reported by Ars Technica And Reuterscompanies of the caliber of OpenAI and Google, but also Microsoft, Meta, Amazon, Anthropic and Inflection are involved, they have promised the Biden administration that they will take concrete steps to make artificial intelligence systems as secure as possible.
We have already indicated on these pages the main issues relating to the agreement between Big Tech and the White House, but it is good to highlight one important point individually, namely the one also highlighted by the sources. Simply put, what is referred to is the introduction of “watermarks” for AI-generated text, video, audio, and imagesso as to try to “put an end” to the plague of content generated in this way “without anyone knowing”.
The goal is obviously to thwart deepfakes and disinformation, aiming to create a “shared system” from the seven companies mentioned above that manages to “clarify” what content has been created using AI. For now, there aren’t so many details on how everything will work, but it is clear that what was outlined on July 21, 2023 represents a significant “declaration of intent”.
For more details, you may in any case also want to take a look at what is indicated directly on the OpenAI official blogbut there are communications in this regard, for example, also from Microsoft And Google (in the latter case reference is made, among other things, to the use of metadata). For the rest, it is worth noting that BigG defined everything as a big step “In the Bringing the industry together to ensure AI helps everyone“.
