Meta-owned WhatsApp’s artificial intelligence (AI) feature that allows users produce stickers based on their search displays a picture of a gun or a boy with a gun when prompted with the terms “Palestinian”, “Palestine” or “Muslim boy Palestinian”, media reported.
When prompted for “Israeli boy”, the AI-generated cartoons of children playing soccer and reading, while for “Israel army” it created drawings of soldiers smiling and praying with no guns involved, according to The Guardian.
The discovery comes as Meta has faced criticism from Instagram and Facebook users who are posting “pro-Palestinian” content.
As the Israeli bombardment of Gaza continues, users claim Meta is enforcing their moderation policies in a biased manner, amounting to censorship, the report mentioned.
However, a Meta spokesperson said the company was aware of the issue and addressing it. “As we said when we launched the feature, the models could return inaccurate or inappropriate outputs as with all generative AI systems. We’ll continue to improve these features as they evolve and more people share their feedback,” Meta spokesperson Kevin McAlister was quoted as saying.
Users also reported that Instagram translated “Palestinian” followed by the phrase “Praise be to Allah” in the Arabic language to “Palestinian terrorist” on multiple occasions.
The company later apologised for that by calling it a “glitch”.
Last month, European Commissioner Thierry Breton wrote to Meta CEO Mark Zuckerberg, telling him to remove pro-Hamas content across his platforms and “be very vigilant” or else it could put the company in violation of new EU regulations.
(With inputs from IANS)
