School AI Blocks and the Personal-Device Homework Rule

A school AI block and a household homework rule answer different questions. Ask which tools the assignment permits, then agree what stays available on the personal device, what pauses, and how to request a temporary exception. A home rule should respect the school's assignment rules, not provide a way around them.

What the new survey actually measured

Common Sense Media's August 2026 report, Teens in the AI Era: Schoolwork and Skills That Matter, surveyed 1,017 US teenagers ages 13–17. NORC fielded it from 30 April to 14 May 2026, combining probability and nonprobability samples and weighting the results to population benchmarks. The report gives an overall margin of error of ±4.3 percentage points at 95% confidence; subgroup uncertainty can be larger, and the hybrid design matters when interpreting that figure.

The questionnaire counted schoolwork both in class and outside class. These are self-reports, not device logs or a homework-only study. The report does not establish where or when a switch to a personal device occurred. Read the primary report.

The 59% is not a percentage of all teenagers, or even of all AI-for-schoolwork users. It is a percentage of the group reporting a school block. Nor are the 63% and 77% groups mutually exclusive: a teen can use AI differently on different tasks.

Getting an answer also covers several reported behaviours. Among users, 35% said they fix or improve an AI answer, 31% rewrite it to sound like them, and 25% use it as-is. Respondents could select more than one. None of these percentages establishes an effect on grades or learning.

This is household device-planning guidance, not medical or legal advice. It does not claim that allowing or blocking AI improves grades or concentration.

Keep chatbot and broader AI figures separate

Pew Research Center asked 1,458 US teens ages 13–17 a different question in September–October 2025: whether they had used chatbots for schoolwork. In its February 2026 report, 54% said they had. Ten percent reported doing all or most of their schoolwork with chatbot help, 21% some, and 23% a little; 45% had not used chatbots that way. Pew's report measures chatbot use, not the broader collection of AI tools in the Common Sense schoolwork study.

The Common Sense census of 9- to 17-year-olds is a third dataset. Its March 2026 survey of 1,204 young people found that 86% had used or interacted with AI, including 81% of 9–12s and 92% of 16–17s. Those are overall AI-interaction figures, not homework percentages. Read the census.

Different ages, field dates, tools, and question wording mean these figures are not a simple trend from 54% to 70% to 86%.

School controls and household rules have different jobs

Figure 9 of the schoolwork report shows personal-device access among some users who encountered a school block. It does not show that school bans fail, that the switch happened at home, or that a particular household control would have prevented it.

The practical inference for families is narrower: do not assume a school network filter describes every rule for a personal device. Ask the school about the assignment and make the household's device plan explicit.

The US Department of Education's August 2026 education-technology and screen-time guidance distinguishes educational and recreational technology and recognises state and local decision-making. That is school guidance, not evidence that a home routine improves homework outcomes.

Rule clarity needs a conversation. In the Common Sense schoolwork survey, 35% did not understand their school's AI rules or were unsure, and 28% said teacher rules varied a lot. Forty-four percent said a teacher had discussed when to use AI for schoolwork, 30% had discussed using it safely, and 27% had discussed what AI is and how it works. These figures describe reported discussions, not a test of what students learned.

If the student cannot explain the assignment's rule, ask the teacher. Our guide to school AI policies and home rules offers a related starting point.

A household rule is not a bypass guide

The personal-device finding is a reason to ask about access, not to list workarounds. Try a sentence the student and supervisor can both repeat:

During the homework window, these named study tools stay available. These apps and sites are paused. If the assignment needs a paused or unclear tool, we check the school rule and request a named, time-limited exception.

Google's official Family Link documentation describes website filters in Chrome and some Android apps, including some WebView-based filtering, and controls for blocking, limiting, or allowing apps on supported supervised devices. System or essential apps have exceptions. Check the documented device and app requirements; this is not a promise of coverage across every browser or app.

Those pages are vendor documentation about controls, not independent evidence that the controls improve homework.

A four-part homework rule, including AI

Write the agreement on paper first, then match it to the tools you use. This is a suggested planning framework, not an intervention tested by the surveys above.

1. Essential access

Identify caregiver calls and messages, required accessibility tools, and agreed health or safety apps. Plan essential access separately from recreational limits, and check on the actual device that those functions remain reachable during the study window.

2. Study tools for this assignment

Name the classroom portal, document editor, calculator, and research tools the task needs. Include an AI tool only when both the assignment rule and the household rule allow it, with a specific job such as brainstorming an outline rather than producing the submitted essay.

If the assignment does not permit AI, using a personal phone does not change that rule. Ask the teacher to resolve unclear or conflicting requirements, including required accessibility support.

3. A pause list

Agree which feeds, games, short-video apps, and chatbots are not needed for this task. A pause can apply to the study window without becoming an all-day rule. Keep the distinction about the agreed use, not simply whether an app has an AI feature.

4. A temporary-exception process

When an allowed, required tool is paused, ask for a specific exception: which app or site, why it is needed, and when access ends. Review an ongoing need separately instead of silently turning a temporary request into permanent access.

Age can inform the conversation, but a survey does not prescribe the right rule. Pew's 2024 parent–teen survey found that 62% of parents of 13- to 14-year-olds reported phone-time limits, compared with 37% of parents of 15- to 17-year-olds. It was fielded in autumn 2023 and measured phone-time limits, not AI rules. This is a cross-sectional age difference, not proof of how any one family's rules should change. Pew's screen-time report.

The American Academy of Pediatrics recommends revisiting a family media plan, including around a new school year, and adapting it as children mature. Include adults in the agreement. Treat the AI section as part of that review rather than a permanent yes-or-no switch.

What the evidence does not establish

• These surveys do not establish that school AI bans or household app limits cause better homework.
• They do not establish that using or avoiding AI causes higher or lower grades.
• The 38% of AI-for-schoolwork users who agreed that AI leads them to generate fewer of their own ideas, and the 39% who felt they missed learning opportunities, reported perceptions. These are not diagnoses or measured learning outcomes.
• The schoolwork survey does not evaluate StudyLumen or show that any app prevents every workaround.

Keep questions about required learning supports with the student, caregivers, and school. A generic device plan should not be treated as a substitute for individual guidance.

Where StudyLumen fits

The agreement above can work on paper. StudyLumen is an option for putting agreed study boundaries into a supervisor-assisted routine on a supervised Android device:

• Scheduled study routines: choose recurring days and local start and end times.
• One selected rule set per routine: apply the chosen app rules/content profile. Website filtering is optional, not an automatic part of every routine.
• Temporary access requests: a student can request access; approved access expires according to the authoritative grant end time, rather than becoming a permanent change to the rule set.
• QR or manual enrollment: link the supervised device to its supervisor without requiring the supervised user to create a separate account.

This is a household-planning use case, not a result demonstrated by the survey. StudyLumen does not replace school policy, detect AI cheating, guarantee grades or concentration, or promise to block every workaround. It does not itself silence notifications; use the device's notification or Do Not Disturb controls for that.

Start with the assignment rule. Then agree the study window, named tools, pause list, and exception process before configuring the supervised device.

Sources

1. Common Sense Media / NORC: Teens in the AI Era—Schoolwork and Skills That Matter, primary report, 18 August 2026. Primary authority for the schoolwork percentages and denominators.
2. Common Sense Media: schoolwork-AI research announcement, 18 August 2026. Context; the article uses the report's measures rather than inferring causal outcomes from its headline.
3. Common Sense Media: report landing page, 18 August 2026. For the teacher-discussion percentages, use the primary PDF.
4. Common Sense Media Census: AI Use by Tweens and Teens, 2026. Broader AI interaction among ages 9–17, not a homework-use measure.
5. Pew Research Center: How Teens Use and View AI, 24 February 2026. Chatbot questions fielded in September–October 2025.
6. Pew Research Center: How Teens and Parents Approach Screen Time, 11 March 2024. Phone-time rules, fielded in autumn 2023.
7. Google Family Link: understand supervision and website filters, checked 4 September 2026.
8. Google Family Link: manage apps on Google Play, checked 4 September 2026.
9. US Department of Education: education technology and screen-time guidance, 20 August 2026. School guidance, not a household outcome study.
10. AAP / HealthyChildren: How to Make a Family Media Use Plan, updated 20 January 2026; checked 4 September 2026.