Meta has introduced new protections for teenage users who discuss sensitive topics such as suicide and self-harm with its artificial intelligence.
But some of the new measures won’t be available in Ireland until later in the year, Meta said.
The new supports include alerting parents if chats with Meta AI, which is built into WhatsApp, Instagram and Facebook, suggest the teen is considering self-harm or suicide. Parents can also choose to limit content for AI experiences for younger users.
The alerts, which are only available on accounts that are supervised by a parent or carer, are already live in the US, UK, Canada and Australia, with the rest of the world, including Ireland, getting them by the end of the year.
READ MORE
Chats that are flagged by AI as potentially concerning will be manually reviewed before alerts are sent, but Meta said if it was ambiguous, the platform would “err on the side of caution” and alert parents.
“While that means we may sometimes notify parents when there may not be real cause for concern, we feel this is the right starting point, and we’ll continue to monitor to help make sure we’re in the right place,” the company said in a blog post.
Will the State’s plan for the Carlton Cinema site revitalise O’Connell Street?
The new measures come after Meta worked with teen mental health specialists to improve how the AI interacts with teenagers on sensitive topics such as suicide and self-harm.
The social network already directs teens in need of support to contact crisis support helplines and trusted adults. Instagram also alerts parents if a teen is repeatedly searching terms related to suicide and self harm within a short space of time. Supervising parents can also see the topics teen users have asked Meta AI about in the past week.
Meta has invested heavily in developing its AI features for its social platforms, spending aggressively on talent and infrastructure as companies vie for dominance in the growing sector.
