The Guardian revealed that the WhatsApp feature, which generates images in response to user searches, displays an image of a gun or a boy carrying it when requested with the terms “Palestinian,” “Palestine,” or “Palestinian Muslim boy.”
The search results varied when tested by different users, but the Guardian verified through screenshots and its own tests that different stickers depicting weapons appeared in these three search results.
Search queries using the term “Israeli boy” returned cartoons of children playing soccer and reading. In response to a request to use the "Israeli Army", the artificial intelligence created drawings of soldiers smiling and praying, without using weapons.
A person familiar with the discussions said Meta employees reported the issue as it escalated internally.
The WhatsApp application, owned by Meta, allows users to try creating special AI images to “create a sticker.” This feature prompts users to “turn ideas into posters using artificial intelligence.”
One user shared screenshots of a search for “Palestinian” that resulted in a different image of a man holding a gun.
This discovery comes at a time when Meta has been criticized by many Instagram and Facebook users who publish content supportive of the Palestinians. As Israel's bombing of Gaza continues, users say Meta is implementing its moderation policies in a biased manner, a practice they say amounts to censorship.
They report being hidden from other users without explanation, and say they have seen a sharp decline in engagement with their posts.
Meta previously said in a statement: “It is never our intention to suppress a particular community or viewpoint,” but due to “large amounts of flagged content” surrounding the ongoing conflict, “content that does not violate our policies may be removed in error.”
Meta spokesman Kevin McAllister said the company was aware of the issue and was addressing it: “As we said when we launched the feature, models can return inaccurate or inappropriate output as is the case with all generative AI systems. We will continue to improve these features as they evolve.” .
A Russian-Saudi agreement in the field of artificial intelligence
King Abdulaziz University (Saudi Arabia) and Russia's South Ural State University agreed to improve and jointly use the system for monitoring pollutant emissions from transportation at road intersections.
The press office of South Ural State University indicated that the project was designed and developed at this Russian university, and the work will be carried out within the framework of an international grant.
The talk is about the AIMS-Eco system, which uses artificial intelligence to analyze emissions plumes left by different types of cars and vehicles at intersections.
Monitoring does not require complex measuring tools, as the system is built on the use of communications and meteorological infrastructure located in cities, and it allows remote access and service to digital environmental monitoring centers anywhere in the world. In addition, Russian scientists have “taught” artificial intelligence to predict emissions taking into account meteorological forecasts.
Vladimir Shepilov, associate professor at the Department of Automobile Transport at the Russian University, emphasized that “foreign methods currently used to determine the amount of emissions on roads work with the help of statistical data. But this approach does not allow managing environmental risks in real time. It is known that changes in temperature Or pressure or wind during the day, can significantly change emissions. It is the continuous analysis of this information that allows us to influence traffic at the right time to reduce emissions harmful to people. The new system has been practically tested in some Russian cities - Perm, Chelyabinsk ", Magnitogorsk, and Petersburg, and he can successfully complement the achievements of scientists from Saudi Arabia."
The Russian expert said: “Our colleagues from Saudi Arabia have created a technology that allows cars to transmit information during movement by transmitting information to the outside about their location and the category of emissions they emit. The integration and integration of this technology with our platform will significantly expand the capabilities of both. The fact is that our cameras now do not see “Except the situation at the intersection, but this is not enough to analyze the situation on part of the road network for a long period.”