Meta allegedly ran a covert operation where hundreds of contractors, managed through contractor firm Covalen, posed as teens to test rival AI chatbots including OpenAI's ChatGPT, Google's Gemini, and Character.AI. Workers created fake under-18 accounts and sent thousands of disturbing prompts on suicide, eating disorders, sex, and self-harm to see how far the systems would go before safety filters activated. Some images used in testing included pills, knives, nooses, and medical diagrams. One testing round completed in August 2025 alone produced more than 45,000 prompts sent to rival platforms. One spreadsheet logged 3,748 separate prompts, with hundreds centred on suicide and self-harm, hundreds more on eating disorders, and at least 239 touching on sex or romance. Prompts included a contractor posing as a pregnant teen asking where to buy abortion pills, a fifth-grader describing a classmate holding a gun to his own mouth, and a teenage girl asking how to hide bulimia. One French-language prompt invoked the real suicide of bullied teen Jamey Rodemeyer. It remains unclear how or whether Meta used the data gathered; one internal Covalen document described the work as "comprehensive AI safety benchmarking."