Twenty digital giants vowed on Friday to crack down on artificial intelligence-generated content designed to mislead voters ahead of a major global election in 2024, AFP reported.

The companies, which include Meta, Microsoft, Google, OpenAI, TikTok and X, are committing to “deploy technologies to counter malicious content generated by artificial intelligence,” in a text released on the sidelines of the Munich Security Conference (MSC).

In particular, they pledge to “work on tools” to “detect” deceptive content created by artificial intelligence so that users can identify it as such and control it.

One idea to achieve this goal would be to apply a “digital signature” invisible to the naked eye but detectable by the device to videos generated by the AI ​​tools developed by these companies.

“All solutions have limitations,” warn the signatories, which include Adobe, LinkedIn, Amazon and IBM.

“We have a responsibility to ensure that these tools are not weaponized in elections,” explained Brad Smith, executive vice president of Microsoft, which invested in OpenAI, creator of ChatGPT.

So-called generative artificial intelligence makes it possible to create text, images, audio and video files by simply asking in everyday language, so that documents created out of the blue can be presented as authentic.

Using archives, the technology can create “deepfakes,” documents that show a person said or did something that never happened.

Digital giants are under pressure to act against the spread of such content as major elections are scheduled around the world in 2024, including in the European Union, the United States, the United Kingdom, India and Russia.

On the same topic:

• OpenAI has developed a technology that allows you to create high-quality videos starting with text-based instructions