Meta implemented temporary changes to its AI chatbot to provide age-appropriate experiences for teens while developing longer-term measures. The company added guardrails to train AIs not to engage teens on romance, self-harm, suicide, and disordered eating, and to direct teens to expert resources. Teen access to AI characters is limited to a select group intended only for education and creative expression. The changes followed a Reuters report revealing an internal document that allowed romantic conversations with children and prompted a senator to announce an investigation. A nonprofit recommended the chatbot not be used by anyone under 18.
"As we continue to refine our systems, we're adding more guardrails as an extra precaution - including training our AIs not to engage with teens on these topics, but to guide them to expert resources, and limiting teen access to a select group of AI characters for now," Stephanie Otway, a Meta spokesperson, told Business Insider. Aside from romantic discussions, other off-limit topics include self-harm, suicide, and disordered eating, Otway said. The AI characters available to teens would only be for the purposes of education and creative expression, says Otway.
After the initial Reuters report, Sen. Josh Hawley wrote in a letter to CEO Mark Zuckerberg on August 15 that he would launch an investigation into how Meta trains its chatbots to have "sensual" conversations with children. "Only after Meta got CAUGHT did it retract portions of its company doc that deemed it 'permissible for chatbots to flirt and engage in romantic roleplay with children," Hawley wrote in an online statement.
Collection
[
|
...
]