Behind algorithms, there are mainly men, and with them, their prejudices. If they are sexist, AI will be equally, or even more so... According to a UNESCO study, generative artificial intelligence perpetuates gender bias. 

Globally, women represent only 22% of team members working in artificial intelligence. "This was not always the case, it was a very feminine profession before the arrival of the microcomputer in the 1980s and it was only from the moment that the sector became more prestigious than men's. "are massively engulfed there", regrets Laure Salmona. 

Indeed, women lack references and models in the sector to be able to identify and think of themselves only as artificial intelligence engineers. To address this, UNESCO launched the Women's Network for AI, which offers tools to ensure the inclusion and empowerment of women at every stage of the development of artificial intelligence. The stakes are high: over the next decade, programming and AI professions will represent the majority of job creations with, as a result, positions of responsibility and therefore a reduction in inequalities.  

What regulation?

The supervision of artificial intelligence has already begun in Europe, with the AI ​​Act. So much so that it is the inhabitants of the Old Continent who should very soon be the best protected. The 27 must definitively approve it in April. But the contours remain vague and the sector is evolving so quickly that it is very likely that politicians are always one step behind...

Any woman can today appear virtually in a pornographic photo or video without ever having participated in it, without even having given her consent. This is what we call pornographic deepfakes! They are just a few clicks away on any search engine, only take about twenty minutes and cost nothing. But they ruin the lives of many women, whether they are famous or not, adults or teenagers and even children.  

96% of deepfakes are of a sexual nature and 99% of them target women. “The parameters are suitable for undressing women, not men…”, underlines Laure Salmona. To the question “Are deepfakes useful or do they represent more risks than benefits”, the cyberharassment specialist seems to have the answer. What if you asked an AI? Anyway, when she doesn't know, she invents. Like us finally!  

The France 24 summary of the week

invites you to look back at the news that marked the week

I subscribe

Take international news everywhere with you! Download the France 24 application