On November 12, 2026, RAND published a report with a finding that flips the usual "AI divide" story. A year ago, teens in higher-income and lower-income families used AI for homework at exactly the same rate. Now they do not. In spring 2026, 57 percent of teens in families earning $150,000 or more used AI for homework, against 43 percent in families earning under $50,000. A year earlier, both groups sat at 39 percent. (Sources: RAND report RR-A4742-5, RAND press release)
Quick summary: RAND (funded by the Gates Foundation) surveyed about 1,000 parents and 1,800 teens aged 12 to 18. Parent AI use rose from 54% to 72% in 15 months, but the income gap among parents widened from 15 to 21 points, 84% for $150K+ families versus 63% under $50K. Teens whose parents often use AI are far more likely to use it for homework (63% vs 40% for parents who never use it). The authors say parental support is the strongest predictor of teen AI use after age. This is survey data and shows association, not proof of cause.
The numbers, in plain English
The report, by Heather L. Schwartz and Melissa Kay Diliberti, draws on RAND's American Parents Panel and American Youth Panel. Parent surveys ran in February 2025 and May 2026, and teen surveys in May 2025 and May 2026. A smaller matched set links 533 teens to 427 parents in the same households. All figures are self-reported. (Source: RAND)
| Measure (RAND, 2026) | Higher-income ($150K+) | Lower-income (under $50K) | Gap |
|---|---|---|---|
| Teens using AI for homework | 57% (39% a year earlier) | 43% (39% a year earlier) | 14 points, up from 0 |
| Parents using AI for work or personal use | 84% | 63% | 21 points, up from 15 over 15 months |
| Parents encouraging AI for schoolwork | 67% | 45% | 22 points |
The gap among teens is 14 points, our arithmetic from RAND's 57 and 43. Overall parent AI use climbed from 54 percent in 2025 to 72 percent in 2026, so everyone is moving up. The wealthiest parents are just moving faster. The report also notes that 17 percent of the highest-income parents encouraged their children to use AI "wherever possible within school rules," even though most parents urged restrained use. (Source: RAND report)
Why RAND points at parents
- Parents who often use AI: 63% of their teens used it for homework in at least some classes.
- Parents who never use AI: 40% of their teens did.
- In the 533-teen matched subset, 81% of teens who perceived parental encouragement used AI for homework, versus 38% who perceived discouragement.
- Per the report, a teen's perception of parental support was more strongly linked to their own AI use than any other factor examined except age. Sex, family income and parents' views of AI harms all ranked lower.
Schwartz, the lead author, said "Parents have as vital a role as schools in helping teens navigate AI," and "Teens need guidance on when AI can support learning and when it risks doing the thinking for them." The authors also write that the uneven use "potentially places at risk the academic performance of teens in higher-income families." Read that carefully: RAND is not saying more AI use means better grades. It is flagging that heavier use could cut either way. (Sources: RAND press release, RAND report)
This does not match every other survey, and that is fine
Pew Research's February 2026 survey of 1,458 teens (fielded September 25 to October 9, 2025) measured something different: the share who do all or most of their schoolwork with chatbots. About 20 percent of teens in households under $30,000 said so, as did a similar share in the $30,000 to $75,000 band, versus 7 percent in higher-earning households. Pew also found roughly 60 percent of Black and Hispanic teens had used chatbots for schoolwork, against about half of White teens. (Source: Pew Research Center)
| Study | What it measures | Headline finding |
|---|---|---|
| RAND, Nov. 2026 | Teens using AI for homework in some classes | 57% higher-income vs 43% lower-income, from 39% each in 2025 |
| Pew, Feb. 2026 | Teens doing all or most schoolwork with chatbot help | About 20% in households under $75K vs 7% in higher-earning households |
| RAND, Mar. 2026 | Students 12 to 29 using AI for homework | 48% to 62% from May to December 2025; 67% say AI harms critical thinking |
These are not contradictions. RAND counts any homework use, which rewards families that coach their kids on how to use AI as a tool. Pew counts heavy reliance, which is more common where kids have less adult help at home. A plausible reading, and it is our interpretation rather than either study's conclusion, is that higher-income teens are using AI more often as a helper, while lower-income teens who use it more heavily may be leaning on it as a substitute. Neither report tests that directly. (Sources: RAND, March 2026, Pew)
What parents can do, whatever your income
1. Ask about the how, not just the whether
RAND's lead author says teens need guidance on when AI supports learning and when it replaces thinking. A useful dinner-table question: "Show me one thing AI explained to you this week that you now understand without it."
2. Know the school's rule
RAND's March survey found only about one-third of students said their school had a schoolwide AI policy, and many said rules vary by teacher. About 80 percent did not consider using AI to understand an assignment to be cheating, while 45 percent considered getting direct answers cheating. Ask your teen's teachers what is allowed on each assignment. (Source: RAND, March 2026)
3. Use it yourself, once, with your teen watching
The RAND data suggests teens notice parents' attitudes and habits. You do not need to be an expert. Try a free tool together on a real problem, then check the answer against the textbook. That models both use and verification.
A homework rule that works at any income level: AI can explain, quiz, and check, but the final answer and the writing come from the student. If the AI gave the answer, ask the student to teach it back without looking.
Limits of the study
- Self-reported surveys. Teens and parents may under or overstate use.
- Association, not causation. Parental encouragement and teen use move together, but RAND does not prove one drives the other.
- Only the top and bottom income groups are reported in the press materials. We did not find middle-income figures.
- No information in the coverage we reviewed on which AI tools or which subjects. The gap says nothing about quality of use.
Not advice: this post summarizes survey research and is not parenting, educational or financial advice. Decisions about your child's AI use should reflect your school's policies and your family's circumstances.
A note on cost and access
If the RAND finding has a lesson for the product side, it is that families who feel less confident need low-friction, clearly explained tools. LumiChats aims to put several AI helpers in one place so a student can compare answers and keep study notes together without juggling subscriptions. Whichever service you pick, sit with your teen for the first session and agree on the ground rules.
01What exactly did RAND find?
In spring 2026, 57% of teens in families earning $150,000 or more used AI for homework, versus 43% in families earning under $50,000. Both groups were at 39% a year earlier.
02Is family income the cause of the gap?
RAND points to parents. Parent AI use and encouragement differ by income, and teens' perception of parental support was the strongest predictor of their AI use after age. It is an association, not proof of cause.
03How many families were surveyed?
About 1,000 parents and 1,800 teens aged 12 to 18. A matched subset of 533 teens and 427 parents allowed within-family comparisons.
04Does more AI use mean better grades?
RAND does not say so. The authors say uneven AI use potentially places at risk the academic performance of teens in higher-income families, so heavier use may not help.
05Why does Pew show lower-income teens relying more on AI?
Pew measured doing all or most schoolwork with chatbots (about 20% under $75K vs 7% higher), while RAND measured any homework use. They capture different behaviors.
