Department of Education AI Guidance 2026: Measure Learning

The federal government finally said the thing out loud: the question is not whether a child is on a screen, it is whether the screen is teaching them anything. The Department of Education AI guidance 2026, issued as a Dear Colleague Letter on August 20, 2026, urges states and districts to judge education technology by “its impact on learning, not simply whether they happen on a screen.” It creates no new rules. What it does create is a list of five questions every ed-tech product should be able to answer, and those five questions work exactly as well from a parent as from a superintendent.

Key Takeaways

• The Dear Colleague Letter, dated August 20, 2026, urges states and districts to evaluate AI and ed-tech by measurable student learning outcomes rather than screen time. It creates no new federal requirements.
• Assistant Secretary of Elementary and Secondary Education Kirsten Baesler wrote it; the guidance distinguishes recreational screen use from technology aligned to clear instructional goals.
• The five questions every product should answer: what learning problem does it solve, when should it be used, for whom, for how long, and what evidence demonstrates improved student learning.
• Districts are told to weigh effectiveness, cost, usability, accessibility, interoperability, implementation support, and educator experience; providers are told to offer evaluations, transparent reporting, and implementation support.
• Baesler’s framing: “The same tools that can distract during recreational use can also expand access to advanced courses, support students with disabilities, and help close learning gaps, when used wisely.”

What the Department of Education AI guidance 2026 says, and doesn’t

A Dear Colleague Letter is how the Department of Education communicates policy interpretation and expectations to states, districts, and institutions. It is persuasion with a federal letterhead, not regulation.

The Department’s announcement on August 20, 2026 lays out the central shift: focus on instructional value over recreational engagement, and let “learning impact” guide selection and review. Baesler’s quote is the balanced version of the argument, which is what makes it citable in either direction: “The same tools that can distract during recreational use can also expand access to advanced courses, support students with disabilities, and help close learning gaps, when used wisely.”

The five questions the guidance says every ed-tech product should address:

1. What learning problem does it solve?
2. When should it be used?
3. For whom should it be used?
4. For how long should it be used?
5. What evidence demonstrates improved student learning?

Districts are directed to consider effectiveness, cost, usability, accessibility, interoperability, implementation support, and educator experience. Providers are asked to build evidence-based tools that strengthen instruction, minimize unnecessary screen time, offer implementation support, and be transparent about design and data use.

What it does not do: create binding rules, set screen-time limits, or require anything of a district that isn’t already required. EdSurge characterized it as long-awaited guidance with implementation deferred to states. That is a fair reading, and it’s why the five questions matter more than the letter’s legal force.

The five questions, turned into things to ask

The letter’s metric	The question to ask your school	What a solid answer sounds like	What a weak answer sounds like
What learning problem does it solve?	”What specific skill is this tool supposed to build?"	"Fluency in multi-digit multiplication; they plateau there every year"	"It’s adaptive and personalized”
When should it be used?	”At what point in the lesson does a student open this?"	"After direct instruction, during independent practice"	"Whenever they finish early”
For whom?	”Which students use it, and who decided?"	"Students scoring below benchmark on the fall screener"	"Everyone gets it”
For how long?	”How many minutes per session and per week?"	"15 minutes, twice a week, capped"	"Until they finish the module”
What evidence?	”What evidence do we have that it improves learning here?”	District pilot data, a published efficacy study, or an honest “we’re measuring this year”	A vendor case study
Effectiveness and cost	”What did it cost and what did it replace?”	A per-student figure and a named displaced activity	”It came with the bundle”
Accessibility	”Does it work for a student with an IEP or in a second language?”	Specific accommodations named	”It’s accessible”

The fifth row is where most conversations end quickly, and that’s informative in itself. “We’re measuring it this year” is a perfectly good answer. “There’s a study” with no citation is not.

Does the evidence back the outcomes-over-screen-time frame?

Mostly yes, and the largest dataset says so in an uncomfortable way for both camps.

PISA 2025, released September 8, 2026, found that 46% of 15-year-olds across OECD countries use AI chatbots at least weekly for learning, and that the pattern of use, not the amount of screen exposure, tracked with scores. Students who used AI daily to summarize assigned reading scored nearly 30 points below non-users in science, roughly a year of schooling. Students in the moderate band scored at or above non-users, and for the use “to help me learn,” weekly users scored highest. The OECD stresses that these are associations and “may reflect a complex mix of who adopts AI and how they use it.” Secretary-General Mathias Cormann’s line matches the federal letter almost word for word: technology “could strengthen learning only when used purposefully and not as a substitute for attention, effort, and understanding.”

There’s also a positive existence proof for the “it can work” half. A Harvard randomized controlled trial (Kestin et al., 2025, Scientific Reports) found students learned more from a carefully designed AI tutor than from an active-learning class session. One well-designed tool, in one course, with a specific learning problem defined in advance. Which is exactly the shape of evidence the letter asks districts to demand.

The caveats worth keeping. Screen time isn’t irrelevant just because outcomes matter more; several districts moved on both at once, with New York City capping individual-device time at 30 minutes daily in grades 3-5 and 45 minutes in grades 6-8 in the same policy that restricted AI. And outcome measurement is expensive. A district that can’t afford an evaluation will not produce evidence just because a letter asked, which is the practical limit of federal guidance with no funding attached.

What to do with this at your school

Ask the five questions about one tool, not all of them

Pick the platform your child spends the most time in and walk the five questions at a conference. Asking about one tool gets you a real answer; asking about all of them gets you a policy statement. If the tool survives the questions, you’ve learned something reassuring. If it doesn’t, you’ve found a specific thing to raise.

Reframe your own screen-time worry as an outcome worry

This is the letter’s actual contribution to family life. “He’s on the Chromebook 90 minutes a night” is hard to act on. “He’s on the Chromebook 90 minutes a night and can’t multiply two-digit numbers without it” is a sentence a teacher can respond to. Same observation, different lever.

Ask what the tool replaced

Not in the letter’s five questions, but implied by its cost-and-effectiveness framing, and the most diagnostic question available. If an AI reading platform replaced twenty minutes of small-group instruction with a teacher, that’s a trade worth examining. If it replaced a worksheet packet, less so.

Watch for the accessibility answer specifically

The letter names accessibility as a selection criterion and Baesler’s quote calls out supporting students with disabilities. If your child has an IEP or is learning English, ask whether the tool was evaluated for them specifically, not just adopted for everyone. A generic “it’s accessible” is not an answer.

Run the same test at home

Before you add an educational app at home, answer the five questions yourself. Most parents find that the app they were about to install has no learning problem attached to it, which saves money and an argument later. We built a version of this into our one-page home AI policy template.

What not to do

Don’t treat the letter as permission to stop worrying about screen time, and don’t treat it as a mandate your district must follow. It is neither. It creates no new rules and defers implementation to states. Its value is a vocabulary: five questions that turn a vague worry into an answerable request.

What to Watch For Over the Next 3 Months

• Week 4: Ask your district whether it has an ed-tech evaluation process and whether it uses outcome measures. Many districts have a purchasing process and no evaluation process; the letter is aimed squarely at that gap.
• Month 2 red flags: A tool adopted mid-year with no stated learning problem. A teacher who can’t say who decided which students use a platform. Vendor case studies offered where district data was requested. Rising screen minutes with flat assessment results.
• Month 3 self-check: For your child’s most-used platform, can you answer all five of the letter’s questions? If yes, you are ahead of most districts. Watch also for whether your state issues its own implementation guidance, since the letter explicitly leaves implementation to states.

Frequently Asked Questions

Does this letter require my district to do anything?

No. A Dear Colleague Letter communicates expectations and interpretation, not regulation. The August 20, 2026 letter creates no new federal requirements and defers implementation to states, which is why it is most useful to parents as a set of questions rather than as a rule to enforce.

Is the Department saying screen time doesn’t matter?

No. It distinguishes excessive recreational screen use from education technology aligned to clear instructional goals, and asks providers to minimize unnecessary screen time. The argument is that screen minutes alone are the wrong primary metric, not that they are irrelevant.

What’s the single most useful question in the letter?

“What evidence demonstrates improved student learning?” It’s the question vendors are least prepared for and the one that separates a tool with a purpose from a tool with a contract. An honest “we’re measuring it this year” counts as a good answer.

Does research support judging tools by outcomes?

The pattern in PISA 2025 supports it: outcomes tracked with how students used AI rather than how much screen exposure they had, with daily summarizing users scoring nearly 30 points below non-users in science and moderate users scoring at or above them. The OECD calls these associations, not causation.

How does this interact with district AI bans?

It sits alongside them. New York City’s September 2026 moratorium restricts student-facing AI through grade 8 and caps device minutes; the federal letter would ask what learning outcome each remaining tool produces. A district can do both, and several are.

What should I ask if my child has an IEP?

Ask whether the tool was evaluated for accessibility with students like yours, which specific accommodations it supports, and whether a human reviews any output that affects placement or services. Washington, DC’s model policy explicitly bars AI from IEP eligibility decisions; that’s a useful benchmark to cite. We cover it in DC’s stoplight AI policy for schools.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. U.S. Department of Education. (2026, August 20). “U.S. Department of Education Releases Guidance on Responsible Use of Education Technology in the Classroom.” https://www.ed.gov/about/news/press-release/us-department-of-education-releases-guidance-responsible-use-of-education-technology-classroom
2. EdSurge. (2026). “Department of Education Issues Long-Awaited Edtech Guidance for States and Districts.” https://www.edsurge.com/news/department-of-education-issues-long-awaited-edtech-guidance-for-states-and-districts
3. THE Journal. (2026, August 20). “Department of Education Issues New Guidance on Education Technology and Screen Time.” https://thejournal.com/articles/2026/08/20/department-of-education-issues-new-guidance-on-education-technology-and-screen-time.aspx
4. OECD. (2026, September 8). PISA 2025 Results (Volume I). https://www.oecd.org/en/publications/pisa-2025-results-volume-i_73451bc5-en.html ; AI-use findings at https://technode.global/2026/09/08/students-avoiding-ai-for-schoolwork-outscore-peers-in-science-bar-one-use-oecd/
5. Kestin, G., et al. (2025). “AI tutoring outperforms in-class active learning.” Scientific Reports. https://www.nature.com/articles/s41598-025-97652-6
6. Office of the Mayor of New York City. (2026, September 2). “Nation’s Broadest Generative AI Moratorium in Schools.” https://www.nyc.gov/mayors-office/news/2026/09/mayor-mamdani-and-chancellor-samuels-put-students-first-with-nat
7. Office of the State Superintendent of Education, District of Columbia. (2026, September 1). “OSSE Releases AI Model Policy to Guide Responsible Staff Use in Schools.” https://osse.dc.gov/release/osse-releases-ai-model-policy-guide-responsible-staff-use-schools
8. MeriTalk State & Local. (2026). “Education Department Urges Schools to Tie EdTech Use to Learning Outcomes.” https://www.meritalkslg.com/articles/education-department-urges-schools-to-tie-edtech-use-to-learning-outcomes/
9. Common Sense Media. (2026, August 18). Teens in the AI Era: Schoolwork and the Skills That Matter. https://www.commonsensemedia.org/research/teens-in-the-ai-era-schoolwork-and-skills-that-matter