Culture in motion. People with influence.Independent voices · Fresh perspectives

Stories shaping the people shaping the world.

The culture & influence magazinePeople · Ideas · Business · Life

In focus

Teens, Social Media and AI Chatbots: What Pew Found in the US

Ideas

Teens, Social Media and AI Chatbots: What Pew Found in the US

Pew Research Center's autumn 2025 survey of US teens shows how social platforms and AI chatbots now sit side by side in daily life. Here is what the figures say, and what they cannot tell us.

Illustration of a smartphone with chat bubbles on its screen and a red speech bubble with three dots beside it

Two reports from Pew Research Center, a US research organisation, describe how American teenagers use social media and AI chatbots. This article sets out what they found, what the figures do and do not show, and why the findings matter to anyone who makes content for young audiences. Everything here describes the United States; the surveys did not cover the UK or the EU.

What did Pew survey, and when?

Pew’s December 2025 report, “Teens, Social Media and AI Chatbots 2025”, and its February 2026 follow-up, “How Teens Use and View AI”, draw on the same survey. According to Pew, it questioned 1,458 US teens aged 13 to 17, and their parents, between 25 September and 9 October 2025. The survey was online, run through Ipsos’s KnowledgePanel, with teens recruited through their parents. Pew reports a margin of error of 3.3 percentage points for both teens and parents.

The sample is weighted to represent US teens living with their parents. That is a sensible design for national estimates, but it means the results speak for that group and not for teens everywhere.

Which platforms do US teens use?

Pew reports that YouTube is close to universal: roughly nine in ten teens say they use it. TikTok and Instagram each reach about six in ten teens or more. Snapchat is used by 55%, Facebook by 31%, WhatsApp by 24% and X by 16%. Pew says Reddit is used by no more than about one in five.

Platform Share of US teens who say they use it (autumn 2025)
YouTube Roughly nine in ten
TikTok About six in ten or more
Instagram About six in ten or more
Snapchat 55%
Facebook 31%
WhatsApp 24%
X 16%

Pew also gives trend comparisons. Facebook stood at 71% in 2014-15, so the fall to 31% is steep. X was at 33% in 2014-15 and 23% in 2022. WhatsApp rose from 17% in 2022 to 24%. These are Pew’s figures for US teens and say nothing about platform use in other countries.

How often are they online?

According to Pew, 97% of teens use the internet daily and about four in ten say they are online “almost constantly”. Daily use differs by platform: roughly three-quarters use YouTube daily, 61% use TikTok, 55% use Instagram, 46% use Snapchat and 20% use Facebook.

The “almost constantly” measure is narrower. Pew reports 21% for TikTok (up from 16% in 2022), about one in five for YouTube, 12% each for Instagram and Snapchat, and 3% for Facebook. Across the five platforms, 36% of teens use at least one almost constantly. Those are self-reported answers to a survey, not measured screen time, so they show how teens describe their habits rather than how long they actually spend.

How many teens use AI chatbots?

Pew found that 64% of teens say they ever use an AI chatbot, while 36% do not. Roughly three in ten use one daily, including 16% who say several times a day or almost constantly. Illustrative arithmetic: 64% who use one plus 36% who do not accounts for all teens.

ChatGPT is the only chatbot among those Pew measured that a majority of teens use, at 59%. Gemini follows at 23% and Meta AI at 20%. Pew also lists Copilot, Character.ai and Claude, each used by smaller shares.

Use differs by group. Pew reports that 68% of those aged 15 to 17 have used a chatbot against 57% of 13 to 14-year-olds, and that Black and Hispanic teens report higher use than White teens (roughly seven in ten against 58%). Pew also notes that daily chatbot use is higher among Black and Hispanic teens than among White teens.

What are teens using chatbots for?

The February report shows a pattern that is mostly practical rather than social:

  • Searching for information: 57%
  • Help with schoolwork: 54%
  • Fun or entertainment: 47%
  • Summarising articles, books or videos: about four in ten
  • Creating or editing images or videos: about four in ten
  • Getting news: about one in five
  • Casual conversation: 16%
  • Emotional support or advice: 12%

For creators, the image and video line deserves attention. Pew’s wording indicates that a sizeable minority of teens have used chatbots for creative work, although the survey does not say which tools, how often or whether the results were shared publicly. Readers curious about how platforms label such material can read our guide to how YouTube and Meta label AI-generated content.

Is AI becoming part of schoolwork?

Pew asked how much teens rely on chatbots for school. Ten per cent say they do all or most of their schoolwork with them, 21% some, and 23% a little, while 45% have not used them for schoolwork. Adding the groups that have used them gives 54%, which matches the separate 54% who say they get help with schoolwork.

Teens are also candid about misuse. Pew reports that 59% think students at their school use AI to cheat at least somewhat often, and 14% say this happens rarely or never. Among teens who have used chatbots for schoolwork, 76% say cheating happens at least sometimes. That is a perception of others’ behaviour, not a measure of how much cheating occurs.

How do teens expect AI to affect them and society?

Pew asked teens about the next 20 years. For their own lives, 36% expect a positive effect, 15% a negative one and 17% are unsure. For society, 31% expect a positive effect, 26% a negative one and 8% are unsure.

Expected impact over 20 years Positive Negative Unsure
On them personally 36% 15% 17%
On society 31% 26% 8%

Among teens who expect a negative effect, the most common reason was overreliance or a loss of critical thinking and creativity (34%), followed by job loss (about a quarter). Around one in ten cited misinformation or misuse. These are teens’ own explanations, offered in response to an open question.

Do parents and teens see the same picture?

Not quite. Pew reports that about half of parents say their teen uses chatbots, compared with the 64% of teens who say they do. The gap is a reminder that parent-reported and teen-reported data can differ, and that a survey of one group should not be read as a survey of both. It is also relevant to rules built around young audiences; our explainer on young audiences, the Children’s Code and made for kids covers the UK and YouTube frameworks, which are separate from anything in Pew’s US survey.

What does this mean for creators and publishers?

The following is this article’s own reading, not Pew’s conclusion. Three points follow from the figures.

First, attention is concentrated. If YouTube is used by roughly nine in ten US teens and TikTok by about six in ten or more, a creator aiming at that age group is working in a few very large venues. How content reaches teens within them is shaped by ranking systems, which our piece on how recommendation algorithms decide what you see explains.

Second, chatbots are mostly used as tools for finding and making things, so they may compete with search and with explainer content more than with entertainment. The survey does not test that, and it would need separate evidence.

Third, teens are broadly aware of AI. Pew reports that 56% have heard a lot about chatbots and 5% nothing at all. Skills for judging what they see online are therefore part of the picture; Ofcom’s approach is described in what media literacy is and why Ofcom promotes it, which concerns the UK.

Where do the findings stop?

Pew’s data are a single survey wave from autumn 2025, so they are a snapshot and may be out of date as products change quickly. The numbers are self-reported, rounded in places and drawn from US teens living with parents. Differences between demographic groups come from subsamples with wider margins of error than the headline figures. None of this tells us whether chatbots help or harm learning, creativity or wellbeing; the survey records use and opinion, not outcomes.

A snapshot, not a verdict

Pew’s findings describe a generation for whom video platforms are near-universal and chatbots are already routine, mostly for information, school and entertainment. Readers should treat them as a dated, US-only picture and check Pew for newer waves before relying on any figure.

This article is general information based on published research. It is not advice, and platform policies and survey findings change.

Sources: Pew Research Center (Teens, Social Media and AI Chatbots 2025); Pew Research Center (How Teens Use and View AI)