Last Saturday, the 25th, an image of the pope in a white coat, or rather a jacket, was circulating on social media.
However, the image, despite having caused an uproar on the internet, is fake and made by an AI. In this article we explain better about the photo of the pope by Artificial intelligence.
see more
These are the 4 zodiac signs that love solitude the most, according to…
There are some dog breeds considered perfect for people…
The pope's photo was made with Artificial Intelligence
If you accessed social media on March 25th, you probably came across an image of the pope wearing a jacket at some point.
The image caused a huge uproar on social networks, as many Internet users began to praise the supposed modernity and style of Pope Francis.
However, the truth is that the image was fake and was built by an artificial intelligence.
The image was not the only false information disseminated thanks to Artificial Intelligence in recent days, after all, we also had false information about the alleged arrest of the former president of the United States, Donald Trump. Trump.
The artificial intelligence used to build these images is called generative artificial intelligence, after all it is capable of generating new things all the time.
In the case of this false information disclosed, the generated content was images. However, another example of generative artificial intelligence is ChatGPT, which is capable of generating new content through texts.
Generative Artificial Intelligence
There are different kinds of artificial intelligences and the main names in the field are Dall-e (from the same creator as GPT), Stable Diffusion and Midjourney. The image of the pope, for example, was created by Midjourney, which published the image on its official Reddit profile.
Midjourney promises updates with even more realistic results and to generate a new image is very simple, just describe the image you want to generate, for example: “a Siamese cat lying on a white sofa in the style renaissance”.
The image is realistically constructed thanks to the “generative adversarial network” which is nothing more than an artificial intelligence learning model.
It works with a second AI evaluating the generated content, as if one machine were saying to another “it wasn't good, do it again” until it reaches a satisfactory result.
The danger of these images
Although it is possible to identify that the image is made by Artificial Intelligence thanks to some errors that the machine misses unnoticed, such as the distorted hands in the pope's photo, the image conveys a lot of credibility because many people overlook it unnoticed too.
In this way, these images become a danger for becoming a potential source of dissemination of deep fakes.
In contrast, DARPA (the US Defense Advanced Research Projects Agency) has started work on a technology to identify fake images and combat the spread of deep fakes.
