Meta beefs up protections for teens on Messenger and Instagram, as new laws loom

Parents and guardians in the U.S., the U.K., and Canada can now check their teenagers’ privacy and safety settings in Messenger, view their contact lists and get updates when someone is added or deleted, and see who’s messaging them. They also get to see how much time their kids are spending on the messaging platform, though they still can’t read their messages.

On Instagram, parents now get to check how many friends their teen shares with the accounts they follow and are followed by. “This will help parents understand how well their teen knows these accounts, and help prompt offline conversations about those connections,” Meta said in a blog post. Insta’s parental supervision tools are also being updated, so teens who block someone are encouraged to let their parents supervise their accounts.

Instagram’s Quiet Mode, which shuts up notifications for a set period, is also being rolled out globally now, having been launched in the U.S., the U.K., Ireland, Canada, Australia, and New Zealand back in January. Quiet Mode was a Facebook feature that was extended to Insta, but the reverse is also happening: Insta’s Take a Break feature is now crossing over to Facebook, so teens there get a prompt to take some time off after they’ve been surfing Facebook for 20 minutes.

“We’re also exploring a new nudge on Instagram that suggests teens close the app if they are scrolling Reels at night,” the blog post added.

This is all sensible stuff, implemented at a time when lawmakers in the U.S. are preparing to crack down on social media companies’ lackluster protections for their youngest users—there are three competing bills out there. Also, in common with some of that proposed legislation, the Federal Trade Commission wants to stop Meta monetizing the data of its under-18 users. 

Will Meta’s latest changes ward off those threats? Not all of them, certainly—Meta isn’t stopping the monetization of its young users, nor is it instituting age-verification measures. However, some of the changes do specifically undercut the rules that have been proposed. 

For example, the Protecting Kids on Social Media Act and Kids Online Safety Act would both block Meta from using recommendation algorithms on teens, to mitigate the threat of app addiction. Now Meta’s lobbyists can tell legislators that the company is trying to reduce the addictive use of its platforms. They can also say parents are getting new controls to identify harmful online behaviors, as the latter act demands.

It’s probably not enough to mollify a Congress where there’s bipartisan support for clipping social media’s wings, but it may help.

More news below.

Want to send thoughts or suggestions to Data Sheet? Drop a line here.

David Meyer

Data Sheet’s daily news section was written and curated by Andrea Guzman.

NEWSWORTHY

DeepMind teases ChatGPT killer. Engineers at DeepMind are using techniques from its computer program that famously beat the champion player of the board game Go to build a new A.I. chatbot the company says will eclipse OpenAI’s ChatGPT. DeepMind CEO Demis Hassabis told Wired that the new large language model, known as Gemini, will incorporate AlphaGo-type systems and could cost tens or hundreds of millions of dollars, and will take months to develop. Plans for Gemini were first shared at Google’s developer conference last month and come as makers of language models aim to have them perform more tasks on the internet and on computers.

IBM’s big-ticket acquisition binge. Tech giant IBM will buy software company Apptio for $4.6 billion. IBM says this latest acquisition—its seventh this year—will accelerate the advancement of IBM’s IT automation capabilities. In a release, IBM boasted that Apptio partners with companies like Amazon Web Services, Salesforce, and Oracle. The announcement comes as CEO Arvind Krishna has made a push into hybrid-cloud and A.I. businesses; the deal is expected to close in the latter half of 2023.

Big Tech tries to squeeze more money from existing customers. Tech companies are increasingly looking for ways to charge for things that used to be free, hike prices, and have customers hand over more money quickly. Take Netflix’s password-sharing crackdown and the spike in subscriptions it saw afterward, or Meta and Twitter charging users for verification marks. And it’s not just hitting users. The dynamic between software developers and services is also seeing a shift, as Twitter and others install price hikes on their application programming interfaces. It’s a strategy that’s reminiscent of airlines eliminating freebies and charging for extra legroom or checking a bag, except with advertising-based internet services, users have already paid by way of being targeted with ads.

SIGNIFICANT FIGURES

59%

—The share of U.S. internet users who have used ChatGPT, according to a new survey from Bank of America. The survey also found that four out of 10 people say they now use OpenAI’s chatbot, or search chatbots offered by Google and Microsoft, several times a week.

IN CASE YOU MISSED IT

Amazon is building a creator platform by mixing commerce with content. One agency manager says top talent hitting $1 million in commissions, by Alexandra Sternlicht

Reddit’s user revolt has exposed weaknesses in Google search—and the tech giant’s employees are not happy, by Nicholas Gordon

NASA is developing its own ChatGPT-like interface in recall of robotic villain HAL from Kubrick classic ‘2001: A Space Odyssey,’ by Chris Morris

IBM’s HR team saved 12,000 hours in 18 months after using A.I. to automate 280 tasks: ‘We’re spending time on things that matter,’ by Trey Williams

Elon Musk says it would ‘not be legal for me to speculate about a Starlink IPO’—but it’ll make the world’s richest man even richer, by Steve Mollman

BEFORE YOU GO

Generative A.I. is developing drugs. INS018_055, a drug that serves as lung disease therapy, is set to be the first entirely A.I.-discovered and A.I.-designed drug to enter a Phase II clinical trial, according to the company behind the treatment, Insilico Medicine. It comes at a moment when A.I. is being hailed as a method of shortening the time it takes to discover drugs since the tech can take in vast amounts of data to find proteins associated with specific diseases along with molecules that can be turned into medicine.

The Financial Times talked to Insilico founder, scientist Alex Zhavoronkov, who said it’s a major test that Big Pharma should be watching. The company even used A.I. to find patients who had a higher chance of responding to the therapy. Now, 60 people will be taking the drug to determine whether it’s a safe and effective treatment for those with idiopathic pulmonary fibrosis. “For Insilico, it is the moment of truth,” Zhavoronkov told the paper.