Students want to talk about AI: what the Teen Tech Council, new AI safety studies, and chatbot dependency reveal
For Educators · The Social Institute
- California launched the country’s second statewide teen tech council: 20 students from more than 10 school districts now advise state leaders on classroom AI use, deepfakes, and digital well-being, with more teens recruited this fall and trained in spring 2027 to lead student well-being projects.
- Most students use AI, and few talk about it with an adult: Two new Common Sense Media studies found that more than 40% of students haven’t had a conversation about AI safety with a parent or teacher, and only 20% have talked with a trusted adult about AI-generated sexual content.
- Students are the ones flagging AI companion dependency: A Drexel University study of more than 300 Reddit posts from students ages 13 to 17 found patterns researchers link to behavioral addiction, including lost sleep, lower grades, and strained friendships and family relationships.
California just gave students a seat at the table on AI policy
The state’s first Teen Tech Council puts 20 students in the room where technology and AI decisions get made.
The Gist
California launched its first Teen Tech Council, convening 20 students from more than 10 school districts in Sacramento to advise the state on technology, AI, and digital wellness decisions. Working with the youth nonprofit #HalfTheStory, the students met with state leaders to weigh in on classroom AI use, deepfakes, and how students interact with technology. This fall, the state will recruit more teens, then train them in the spring of 2027 to “lead projects that directly improve the wellness of teens,” according to the governor’s office. As big proponents of elevating student voice, we love to see students living through these changes helping decide how the state responds to them.
What to Know
California has been one of the country’s most active states on tech laws for students, pushing proposals to ban social media accounts for kids under 16 and limit minors’ use of AI chatbots. It’s also the second state to try a youth tech council, after New York launched one in 2025, as more than a thousand school districts are currently suing platforms like Meta and Google, arguing their algorithms hooked and harmed teens.
The Teen Tech Council goes a step further than legislation alone. It brings the students most affected by these rules into the room to help write them. Challenges like deepfakes and AI companions hit student well-being and learning directly, and students often notice those effects before adults do. When they sit at the table with educators and officials, the policies that follow reflect what students actually face day to day.
TSI’s Take
Students have real, valuable insight into how AI is shaping their lives, and the schools that tap into that insight are better equipped to build sustainable policies. California’s Teen Tech Council is a great example of what it looks like when student voice is treated as a resource rather than an afterthought.
Educators can bring that same energy into their own schools by:
- Recognizing that students are often ahead of the curve: This is a perfect opportunity to empower students to help lead.
- Asking students what they’re actually experiencing with AI: The best guidelines are ones students helped shape.
- Creating space for student-led conversations about technology: Students who feel heard are more likely to engage responsibly.
Help students become empowered to lead with the #WinAtSocial Lesson, Student voice in school tech and AI policies. In this lesson, students reflect on how their school’s AI policies impact their focus, learning, and school culture, and what they can do to change it.
Two new studies show how students are really using AI, and what adults are missing
Students are taking their most personal questions to chatbots, and most haven’t talked about it with a trusted adult.
The Gist
A pair of new studies from Common Sense Media is painting a clearer picture of how teens are actually using AI, and it goes well beyond homework help. Students are asking chatbots about their bodies, their relationships, and other challenges they might be facing on a day-to-day basis. The studies also found that AI is lowering the barrier to sexually explicit content, with nearly half of teens reporting they’ve encountered AI-generated explicit material, often without even seeking it out. It’s a reminder that students are working through big, personal questions with AI, and that’s exactly where a trusted adult can make the difference.
What to Know
“Using AI is nearly universal among tweens and teens in America,” said Geoffrey Fowler of Common Sense Media’s Youth AI Safety Institute. The research found that teens who report feeling lonelier tend to use AI more frequently for emotional support, and those struggling academically are turning to it for help.
More than 40% of students said they haven’t had a conversation about AI safety with a parent or teacher. Only 20% of teens have talked about AI-generated sexual content with a trusted adult. Fowler noted that families don’t need to be tech experts to start these conversations. The open door matters more than the expertise.
TSI’s Take
Students are navigating some of the most personal parts of their lives with AI, often without anyone they trust knowing about it. That’s not because students don’t want guidance. The guidance simply hasn’t shown up yet.
Educators can help close that gap by:
- Starting the conversation before students feel like they have to figure it out alone: A simple huddle about what students are actually using AI for is more helpful than a printed-out policy.
- Helping students understand what AI can and cannot do: A chatbot can organize your thoughts. It cannot replace the feeling of being genuinely heard by someone who knows and cares about you.
- Making it easy for students to come to a trusted adult: Students who know they have a go-to person are more likely to reach out before a situation escalates.
Help students understand their relationship with chatbots by running the #WinAtSocial Lesson, AI as a thought-partner, where students explore how AI is part of our everyday lives and understand both its benefits and limitations.
Students are noticing when their AI chatbot use becomes a challenge
A Drexel study of 300 Reddit posts found teens describing dependency on AI companions, and asking for help striking a balance.
The Gist
A new Drexel University study analyzed more than 300 Reddit posts from teens ages 13 to 17 who wrote about leaning too heavily on AI companion chatbots like Character.AI, Replika, and Kindroid. What began for many as emotional support or entertainment grew into patterns researchers link to behavioral addiction, including conflict, withdrawal, relapse, and mood modification. Some said their heavy use cost them sleep, hurt their grades, and put a strain on their friendships and family. It’s a sign that tools built to keep students company can quietly start to take over, and many teens are saying they want help spotting when that happens and skills to strike a balance.
What to Know
AI companions are different from the other tech teens have used in the past. They’re built to feel like relationships. They talk back, respond to your feelings, and get more tailored to you over time. According to researcher Matt Namvarpour, “stepping away is not just stopping a habit, it can feel like distancing from something meaningful.”
In about one in four of the posts studied, teens described leaning on the bots to get through hard moments, from loneliness to real well-being struggles. The researchers said designers should build in off-ramps and usage-awareness tools, and they noted that teens are already spotting the challenge and asking for help with it.
TSI’s Take
Teens are posting about their chatbot dependency, which tells us they want to talk about it. They just need a responsible adult they trust and a way to think it through.
Educators can help students find that by:
- Normalizing conversations about how students use AI for connection: Asking “when do you find yourself reaching for a chatbot instead of a person?” opens the door without judgment.
- Helping students identify who their real go-to people are: Students who have trusted adults and strong peer relationships are less likely to fill that gap with a bot.
- Framing this as striking a balance: AI companions can be useful, and students still need guidance on when a tool is helping them and when it’s holding them back from the real connection they actually need.
In The Social Institute’s #WinAtSocial Lesson, The impact of AI on everyday tasks, students discuss the appropriate times to use AI and reflect on how they can use it in their daily lives without becoming overly reliant on it.
Students are already telling us how AI is shaping their focus, their friendships, and their well-being. California is building a council to hear it. Common Sense Media and Drexel researchers had to read studies and Reddit posts to find it. Educators get to hear it first, in the hallway and in the classroom, when students know the door is open. Empowering educators to have those conversations is what #WinAtSocial is built for. Want to see how it looks in your school? Request a demo and we’ll walk you through it.