New rules concerning deepfakes were adopted in the Senate, during the examination of the law “to secure and regulate the digital space”. They bring changes, especially on pornographic deepfakes.
Reinforced fines and better supervised deepfakes. On July 5, 2023, during the Senate review of the “Securing and Regulating the Digital Space” law, new rules concerning fake videos generated by artificial intelligence were adopted.
Jean-Noël Barrot, Minister Delegate in charge of the digital transition and telecommunications, went to the Senate to defend the amendments proposed by the government. He welcomed on Twitter the adoption of these new laws, which strengthen the arsenal of the executive in the fight against deepfakes.
New rules against pornographic deepfakes
To fight against these deepfakes used for harmful purposes, the Senate has just adopted new rules. ” Publishing a deepfake representing the words or the image of a person without their consent, and without it being clearly established that it is a fake will be punished up to 2 years in prison and 45,000€ in ‘fine “said Jean-Noël Barrot.
An amendment specifically targeting pornographic deepfakes was also passed. These videos which, as the Minister reminds us, aim ” almost always women whose voice, face or body are realistically reproduced “, will now be punished” up to 3 years in prison and a fine of €75,000 “.
Deepfakes, these “ hyperrealistic special effects as the Minister says, are fake videos, which depict real or invented people doing things they never did. These videos are made using artificial intelligence, which makes them look very realistic.
Deepfakes were first popularized by their pornographic use, by putting the faces of famous women on the bodies of pornographic actresses in X movies. They are now also used to produce disinformation, especially in the context of war in Ukraine: a fake video of President Zelensky, calling on the Ukrainian army to surrender, circulated in March 2022.
Subscribe to Numerama on Google News to not miss any news!