(Subscribe to our Today's Cache newsletter for a quick snapshot of top 5 tech stories. Click here to subscribe for free.)
In an effort to safeguard its teenage users, Instagram has added a feature that will prevent adults from sending messages to people under 18 who don’t follow them. If an adult tries messaging a teen on the platform, they will receive a note that sending a direct message (DM) to the user is not an option.
The photo-sharing app also said the move is part of Facebook’s larger goal to make its platforms end-to-end encrypted and safer for younger users . Consequently, the feature will be rolled out in Facebook’s Messenger app too.
Instagram has faced criticisms in the past for not doing enough to prevent child abuse on its platform. It was questioned for not age-gating new user sign ups while its parent Facebook had an entry barrier for under-age children. The social network reported over 20 million child abuse images on its platform in 2020, according to the U.S.-based not-for-profit organisation National Center for Missing and Exploited Children (NCMEC).
Just about a year ago, the app drew a date-of-birth boundary for its new users. And prior to setting up age-gating barrier, new users flocked to Instagram without giving information about their age. This boundary-less approach had put the app in the territory of breaching child protection laws in the U.S., particularly Children’s Online Privacy and Protection Act (COPPA). The primary aim of the rule was to give parents control over personally identifiable information (PII) of their under-age children. Violation of COPPA could cost about $43,792 per user.
With AI help
Even after implementing age filter for new registrations, the platform continues to face persistent issue of incorrect or inflated age data. This makes younger users to sign up and stay on in the network. So, Instagram says it is resorting to machine learning to flag potentially inappropriate interactions, improving teen privacy features and DM-ing users with real-time safety information. If a conversation is flagged as inappropriate, teens can choose to end it, or block, report or restrict the adult.
The platform also said it will device ways to restrict potentially harmful adults from appearing on teenage users’ ‘suggested users’ and from Reels and Explore section, the company wrote. Facebook will also send timely prompts to users with public profiles to keep their accounts private.
Using artificial intelligence (AI), Facebook says it will also verify date-of-birth entered by a teen user and block access if they are under 13. The social network is also working on making it difficult for adults who show potentially harmful behaviour on its platform from interacting with teens.
It is not clear how Facebook will use AI to know people’s real age. The company is facing a lawsuit in Illinois for collecting people’s biometric data, including facial recognition scans, without their consent.