with the head of the cow. and that s because it s completely invisible to the human eye. you can only see it if you run it through google s own image generator tool, which is called imagene. and all it says on it is that the image was made by ai. it s got no other information about who made it, where it was made or when it was made. now, this is only a trial at the moment, and deepmind says you have to opt into it. so obviously anyone who s deliberately trying to set out to deceive using an ai image is probably not going to tick that box. but it is an interesting way, perhaps, of looking at how we might protect ai images in the future. in china, for example, you have to watermark or give some suggestion that you re dealing with al content, and they are bringing that in as part of their regulations. and there s every sense that companies are going to have to be transparent about when people are interacting with al. so it is one way of doing it. it s quite robust, deepmind says. if you e
so obviously anyone who is deliberately trying to set out to deceive using ai image is probably not going to tick that box. but it is an interesting way perhaps of looking at how we might protect ai images in the future, in china for example, you have to watermark or give some suggestion that you are dealing with al content. they are bringing that in as part of their regulations. there is every sense that companies are going to have to be transparent about when people are interacting with al. so it is one way of doing it, it is quite robust. deep mine says. if you edit the image, if you change it in any way, that watermark will still remain bitter they want people to play around with it, if you like, to experiment, they want to track it and see whether it is actually good enough to be a tool that we might use, particularly in the fight against the disinformation in the future. talk to me a little bit about that fight, why is it so important deep mine says. that these types of initiativ