We went to EdTech Week with a question: how are EdTech teams answering the new AI questions parents and schools are asking? Across dozens of conversations with founders, product leaders, and district partners, one takeaway stood out. Most teams are making responsible AI choices. Very few have a way to show them.

The questions have changed, and everyone knows it

Is my child's writing training a model? Does the reading app keep recordings of my child's voice? Is it tracking what my child reads? Nearly every team we spoke with is already hearing questions like these from parents and districts.

The stakes are concrete. This school year, New York City and Los Angeles Unified paused student-facing generative AI while they review its use. New York's pause extended even to tools that had previously been approved. Several people we spoke with described that as a turning point: approval once no longer means approval for good.

"Trust with our partners matters more than anything else we do. District and school partners need a way to verify that the tools they use every day maintain the highest standards for privacy and security, not just take our word for it. The Common Sense Seal is the trusted, independent evaluator of that standard." — Arman Jaffer, Founder & CEO, Brisk Teaching

Why good choices still get a "not yet"

The teams we met are doing real work. Many don't train on student data, keep teachers in the loop, and limit what flows to model providers. But from the outside, a good choice no one can see looks exactly like a bad one. When schools can't tell a carefully built AI product from a careless one, "not yet" becomes the safe answer for all of them.

Part of the problem is that the familiar frameworks weren't built for this. FERPA and COPPA set important baselines, but they were written decades before generative AI. They don't ask whether a child's writing trains a model or what happens to a recording of a child's voice. Data privacy agreements and privacy policies can still do the job, as long as they ask and answer those new questions.

States are starting to fill the gap with AI-specific rules, and they came up often in hallway conversations:

  • California will prohibit using K-12 student information to train or develop AI systems starting in July 2027.
  • Oklahoma now requires districts to give parents an annual AI inventory: which tools they use, who provides them, what student data each collects, and how it's shared. Families can opt their children out of student-facing AI, and AI can't be the primary basis for grading, discipline, or placement.

Districts can't meet requirements like Oklahoma's without clear answers from their vendors. And because these rules are arriving state by state, with different requirements and timelines, schools and EdTech companies alike are working to keep up.

Trust has to be visible, and it has to stay current

Even when a district does its homework, the evidence usually lives in a contract parents will never read. Families see the app, and too often that means a chat box with no explanation. A review at the time of purchase isn't enough either. Products ship new AI features every quarter, so reviews have to keep pace.

The teams furthest along treat transparency as part of the product. That often has nothing to do with a new model. It looks like:

  • Clear labels showing when AI is in use
  • Plain-language explanations of what happens to student information
  • Published lines the company won't cross
  • Someone outside the company checking that it all holds up over time

That last point is why we built the Common Sense Privacy Seal. Families, teachers, and schools already turn to Common Sense's privacy ratings when choosing products. The Seal builds on that with an independent review of a product's data and AI practices, including how student information flows into AI features and how clearly that use is explained. With ongoing monitoring, companies must continue to meet Seal standards to keep it.

So what's next after EdTech Week?

New York and Los Angeles have made their call for this school year, and other districts are weighing the same questions. What we heard at EdTech Week suggests the companies that earn lasting trust won't be the ones with the most AI features. They'll be the ones whose responsible choices are easy to see, with clear, verifiable answers ready before anyone asks.

Thank you to everyone who chatted with our team. If we didn't connect in New York, we'd still like to hear how you're approaching these questions. Book a meeting with our team and we'll walk through where your product stands and whether it's eligible for the Common Sense Privacy Seal.