Distinguishing fake images generated by artificial intelligence from genuine images is an increasingly pressing problem, writes The Guardian Interested in trade.
It is said that even experts have serious trouble identifying fake products, at least this was revealed when testing the latest content generators. Providing images with watermarks is no longer a guarantee of anything, because fake images generated by artificial intelligence can also be easily watermarked.
The main reason for the problem is that AI innovations are evolving at a much faster pace than security solutions or procedures of social media providers. These artificially generated images often appear as the first results in Google’s image search engine, so chaos has taken over Google’s search engine as well.
A professor at the University of Pennsylvania explained in a post on Twitter that he believed he had already done so We can’t trust anything on the Internet anymore: counterfeiting has become trivial and impossible to tell the difference.
However, the good news is that in the near future, Google plans to conduct a more comprehensive review of the history of images, through which it can better filter out scams.
