Now what?
Much has been done to stop Fake News. Companies were forced to act.
Much has been done. There is much more to do. In the time of Artificial intelligence, Fake News is even easy to detect. The problem is with Deepfakes…
Meta has implemented a Fake News detection algorithm that is then verified by national teams. At the same time, Meta started financing Fact Checker platforms.
X (Twitter) created Community Notes, a mechanism through which verified users can report Fake News and/or misinformation.
Google gave another highlight to its reverse search tool for images.
The European Union's Digital Services Act (2022) created a regulatory framework for large platforms and which, among other points, holds these platforms responsible for the display and dissemination of Fake News. In other words, platforms are now obliged to take action and remove this online content.
The European Union also launched EUVSDesinf. A platform that identifies and reports narratives used for disinformation.
In March 2023, a photograph went viral on the Internet and was even featured in Vogue Brazil: Pope Francis in a long white padded coat. It was a deepfake.
The fake image was created in an Artificial Intelligence app - Midjourney. The photograph had originally been published in the Reddit community, but was quickly shared and disseminated on various social networks as true.
In 2023, fake images of former US President Donald Trump, being convicted, and in prison, also flooded social media. The photographs were created using text commands, with Artificial Intelligence (AI Midjourney). Many people believed them to be true.
In May 2024, a rare phenomenon occurred in Portugal: it was possible to see northern lights in the country's skies. The news and social media showed images of the aurora borealis in the skies of Lisbon and Porto, however this phenomenon would only be visible in less lit areas, outside of urban centers. The images were not true, but there are even those who claimed to have seen the aurora borealis in the same places shown in the photographs, even if that had never happened.
The images were processed with an artificial intelligence tool and began to be shared on social media as if they were real photographs.
Artificial intelligence engines today have the capacity to generate text, voice, video and moving images. They have the ability to deeply manipulate a video and the words of an interlocutor to the point where we cannot distinguish what is “true” or “false”.
When this happens, we talk about Deepfake.
The evolution of these engines has been exponential. Its impact has been widely discussed in software companies. As well as creating control tools for their application.