This week’s news felt like one long reminder that whether parents asked for it or not, the AI rollout in childhood is no longer pending. It is happening. The federal government just made AI more attractive in education grants. UNESCO is politely reminding the world not to lose its mind and replace actual human judgment with whatever a chatbot spits out. Meanwhile, police are dealing with AI-generated sexual images involving kids, teens are openly talking about getting too attached to companion chatbots, and many parents still do not realize how often their kids are already using AI for schoolwork.

So no, this was not an “AI is coming soon” kind of week. It is here, it is messy, and as usual, the adults expected to sort through it are parents, teachers, and school staff with limited time and regular responsibilities. These stories are about what adults need to be paying attention to before this all becomes classroom normal and home normal without anybody stopping to ask better questions.

If you want to go deeper, start with Why Games Ask for Your Birthday and Why Everything You See Online Is Trying to Get Your Attention.

Now time for the news.

The U.S. Department of Education finalized a rule that formally prioritizes artificial intelligence in its discretionary grant programs. Schools and colleges applying for federal education grants may now get a boost if their proposals help students better understand AI or use it in responsible ways. There is also extra interest in programs that actually teach AI literacy as part of instruction. This does not mean there is a new pot of AI money. It just means AI-related proposals may get more consideration within existing federal grants. The move builds on an April 2025 executive order that told federal agencies to expand AI efforts in K-12 schools.

What parents and teachers need to know: If your school suddenly has a new AI product next semester, this is part of why. The questions worth asking at the next board meeting or back-to-school night are not complicated: Who vetted this tool? What data does it collect on my child? What is the teacher allowed to override? A polite email to the principal asking those three things costs you nothing and tells you a great deal about how seriously the district is treating the moment.

UNESCO put out updated guidance saying AI in education should stay focused on people, not just technology. The group stressed that AI should not make existing gaps worse for students who already have less access to devices, internet, or support. UNESCO also released new frameworks to help policymakers, teachers, and students make smarter and safer decisions about using generative AI in schools, including both the opportunities and the risks.

What parents and teachers need to know: We love to see global leaders remembering that technology should serve people, not the other way around. As the adults in the room, it is our responsibility to make sure these shiny new programs are actually helping our kids grow, not just doing the heavy lifting for them. When your local school introduces a new AI tool, don't be afraid to ask how they are ensuring every student gets fair access, and more importantly, how they plan to keep the human touch central to learning.

The Isle of Man Constabulary went public with an account of how AI-generated deepfakes and manipulated images are reshaping childhood bullying on the island. PC Louise Kennaugh, a school education officer, said some kids have ended up changing schools or even leaving the island after classmates used AI to create fake sexual images of them. Some of the children affected were as young as seven. She also said that when she asks primary school students whether they have seen something online that scared them, about three out of four raise their hands. Right now, the police response is focused more on education than punishment for the children making these images.

What parents and teachers need to know: Although this happened across the pond, the tools those seven-year-olds used are on the same app stores your seven-year-old uses. The children doing the harm and the children being harmed are often sitting in the same classroom, and both need adults who can talk about images, consent, and regret without freezing up. Start the conversation before it becomes a police matter. A child who has heard the phrase “fake image” from you first can be prepared for when they encounter one from a peer.

Police in Richland, Washington said AI-generated child sexual abuse images are rising fast, and the law is struggling to keep up. One big problem is proving whether an image is completely fake or whether it was made from a real child’s photo, because that can determine whether charges are possible. Reports of this kind of material to the National Center for Missing and Exploited Children jumped from the thousands in 2023 to more than a million in 2025. One local case involved a school para-educator who used AI to make explicit images of a real 12-year-old girl using photos from social media. That case helped push Washington state to pass a new law so prosecutors can bring charges even when a specific victim cannot be identified.

What parents and teachers need to know: Two things are true at once here. The first is that this is a hard story to read, particularly if you are a parent who posts pictures of your child. The second is that the answer is not to delete every photo and move to a cabin. The answer is to be selective. Public accounts, high-resolution face-forward images, and school-uniform-identifying photos are the ones worth reconsidering. A private grid for family and a quieter public presence goes a long way. And when you talk to your child's teachers and coaches about photo policies, you are not being the difficult parent, you are being the one dealing with current issues.

Kalamazoo Public Schools offered one of the more grounded stories of the week. After a lengthy vetting process, the district chose an AI tool it said better protects student data and privacy, is using it in limited ways like study support, quiz help, text rewriting, and character-based classroom activities, and is pairing that with parent seminars after families raised concerns about chatbot use at home. So the message was not “AI everywhere, good luck.” It was more like: bring it in carefully, keep adults involved, and talk to families before a home issue turns into a school problem.

What parents and teachers need to know: This is a good approach for more schools to copy. If AI is coming into class, families should not be left to sort out the home side alone after a weird late-night chatbot moment. Put family workshops, plain-language guidance, and a real human contact person right next to the rollout.

A Drexel University study released presented this week at the Association of Computing Machinery's conference on Human Factors in Computing found that more than half of U.S. teens say they use AI companion chatbots regularly, and some are showing signs of getting too attached. Researchers looked at more than 300 Reddit posts from teens ages 13 to 17 who said they felt dependent on Character.AI. Some talked about losing sleep, falling behind in school, and having a harder time staying connected to people offline. What stands out is that these concerns were coming from the teens themselves, not just worried parents. The researchers say companies should build these tools more carefully, with attention to kids’ emotional needs, unhealthy attachment, honest empathy, and making it easier to step away.

What parents and teachers need to know: Teenagers are the ones raising the alarm. They know something is off, and they are writing about it on Reddit because the adults around them do not always know the product exists. The move here is a question: do you have an app you sometimes wish you used less? Let the answer sit. Then say the same is true for you, about something of yours. A kid who feels caught will hide. A kid who feels seen will actually show you the app on their phone.

-Matt Silverman, Digital Producer (Quote from AI for Kids Podcast)

In a commentary for EdSource, USC education professor Morgan Polikoff argued that the data on student AI use is already pointing in a troubling direction. Pew found that more than half of teens are already using AI for schoolwork, and about 1 in 10 said AI is doing all or most of the assignment. Polikoff’s USC survey found the adult awareness gap is still pretty bad: only 7 percent of parents thought their teens were using AI for school multiple times a week, while 27 percent of teens said they were, and he suspects even that number is low. Parents also largely did not know their schools’ AI policies. His takeaway was straightforward: schools need to put more assessment back in the classroom, cut the busywork, and do a much better job helping both students and parents understand the risks.

What parents and teachers need to know: The gap between 7 percent and 27 percent is a lot. It is not that teens are hiding anything especially well. It is that the rest of us have not yet made it a normal thing to ask about. A small habit that changes this: when your teen sits down at the kitchen table to do homework, ask what they are working on, and then ask how they are working on it. Teachers, if it is in your power, a quick “walk me through your thinking” at the start of class does more than any detection tool on the market. (Which we know may or may not work properly.)

New research from Chicago Booth gets at a real tension a lot of parents are feeling right now. On one hand, many worry that too much AI could weaken their kids’ thinking and problem-solving skills. On the other hand, they are even more worried about their kids falling behind if everyone else is already using it. In a study of 2,000 parents, the push to pay for premium AI tools was driven less by excitement about the technology itself and more by the fear that other kids are getting ahead with it.

What parents and teachers need to know: It is completely natural to worry about your child falling behind the curve. We all want to set them up for success. But let's take a collective deep breath, we don't need to hand over unrestricted access to AI just because it feels like everyone else is doing it. It's about finding a healthy balance. Teach them to use these tools as a sounding board or a tutor, not as a crutch to do the thinking for them. Keep an eye on how they problem-solve independently, and remind them that their own brain power is still their most valuable asset.

Tied to the deepfake and AI-image stories this week. Playable at the dinner table, on a car ride, or in a classroom.

What You Need: Nothing but a few people ready to play the game.

How to play

Why it works: This game builds the exact habit kids need with AI: asking one smart follow-up question instead of just trusting something because it sounded confident. They start to see pretty quickly that odd, specific details like “what color was the mug?” can expose a made-up story faster than broad questions like “what happened next?” That same instinct matters online too. It is what helps kids slow down and notice when an image, video, or AI answer feels off.

If there is a throughline to this week’s news, it is that the adults shaping how AI shows up in childhood are usually not the ones building the tools. They are the ones sitting next to a kid at the kitchen table, standing in front of a classroom, writing policy at a school board meeting, or answering the phone after something has already gone sideways. That gives me the most hope. The people closest to the mess are also the people closest to doing something about it.

So if this week’s roundup made you think of someone, send it to them. A teacher. A parent. A grandparent who is still trying to figure out what ChatGPT actually is. That is how this all gets a little less confusing for the rest of us. And if you have a question or want me to cover a certain topic, send me a note. Thanks for all you do for the kids in your life.

Until next Tuesday,

Amber Ivey (AI)