Happy Sunday! 

It’s Cory Turner, correspondent on the team. By the time you’re reading this, I will have spent a butt-numbing number of glorious hours sitting, watching The Lord of the Rings, which is returning to movie theaters for an anniversary that won’t be named since the movies can’t possibly be that old already. Since that would make me even older, already. 

Oh, time.

On the work front though, I spent the past week deep-diving into a new global study on the promise and peril of generative AI use among children and teens. Full disclosure: I am the parent of two teenagers who, like The Lord of the Rings, appear to have gotten much, much older while I was sleeping. This report was a real eye-opener for me – and not just because I’m a parent with skin in the game.

This new report, from the Brookings Institution, explains that free, broadly accessible chatbots are largely sycophantic in design. Which means, they tell us what they think we want to hear (or, rather, what their algorithm predicts we want to hear). The risk this poses to children and teens is considerable. Empathy is a learned skill. A healthy society is full of adults who, as kids, learned how to listen and relate to others who may look, think and feel differently from them. That’s part of the work of school.

But chatbots now offer kids “companionship” with a friend – who won’t disagree with them or challenge them or ask for help. Which helps explain why 1 in 3 teen users say they prefer having serious conversations with chatbots instead of with their fellow humans.

The Brookings report is full of research-driven tidbits like that, which can help make sense of how children and teens think about AI and why, the researchers argue, it’s important for the U.S. and other governments to do more to regulate it.

You can read my full story here.