jour]

Here’s a rewritten article based on your provided text, adhering to your core mission:

AI Image Generator “imagefx” Sparks Controversy with Racist and Stereotypical Content

Marseille, France – A new artificial intelligence tool from Google, known as “ImageFX” (formerly “Imagine 3”), is facing widespread criticism for generating videos that perpetuate harmful stereotypes and racist narratives, especially concerning the city of Marseille and its inhabitants. The tool, which allows users to create videos with sound effects, ambient noises, and dialog, has been deployed in Europe since early July, following its US launch in May 2025.

The controversy centers on videos created using ImageFX that depict exaggerated and negative portrayals of Marseille. One widely shared video features a young woman, presented as an influencer, documenting her “stay in Marseille.” The AI-generated narrative includes her claiming to have had her watch stolen in the “northern neighborhoods” and being assaulted “only twice this morning.” The video further depicts the influencer adopting a hijab and burka, and speaking Arabic, with the accompanying text from the user stating, “Process of integration of an influencer in Marseille. Seriously, we will have to identify these people.” This has led to questions about whether the user or the AI is the primary source of the racist character of the content.

This incident highlights a broader concern about AI’s potential to reinforce existing biases. In another example cited by TV5 Monde, a user requested a “banal experience in Africa” from the AI. The resulting video depicted a white man with a selfie stick and a bottle of water, followed by a black child asking for water. A TV5 Monde journalist commented that such a scene “perpetuates the image of a poor Africa requiring the help of white men.”

This is not the frist instance of an AI tool generating problematic content. In 2016, Microsoft’s conversational AI, Tay, was taken offline within 24 hours of its release due to its output of racist and hateful responses, as well as the dissemination of false data.

The ImageFX tool’s ability to translate simple text descriptions into complex, often audio-visual, scenarios has raised notable ethical questions about the responsibility of AI developers and users in preventing the spread of misinformation and harmful stereotypes.As AI technology becomes more sophisticated and accessible,the need for robust content moderation and ethical guidelines is increasingly apparent.