Pew Study Reveals How Teens Use Social Media and AI Chatbots

The Pew Research Center released a new study on Tuesday highlighting how U.S. teens interact with both social media and AI chatbots. The findings arrive as global debates over teen internet safety continue to grow. Australia, for example, plans to enforce a social media ban for kids under 16 starting Wednesday.

Concerns around teen mental health remain high. While some research shows online communities can offer support, other studies link doomscrolling and excessive screen time to negative mental health outcomes. Last year, the U.S. surgeon general even urged social platforms to add warning labels.

Teens Are Online Constantly, and AI Is Now Part of the Habit

Pew reports that 97% of U.S. teens use the internet daily, and about 40% say they are “almost constantly online.”

Although this is slightly lower than last year’s 46%, it is dramatically higher than a decade ago, when only 24% reported being online nearly nonstop.

AI chatbots are now a core part of that online experience. Pew found that 3 in 10 U.S. teens use AI chatbots every day, and 4% use them almost constantly.

59% use ChatGPT, making it by far the most popular chatbot
23% use Google Gemini
20% use Meta AI
46% use AI chatbots at least several times a week
36% do not use chatbots at all

Pew Study Reveals How Teens Use Social Media and AI Chatbots — Image credit: Pew Research Center

Usage Varies by Race, Age, and Household Income

Pew’s data shows notable demographic differences:

Race and Ethnicity

68% of Black and Hispanic teens use chatbots
58% of white teens use chatbots
Black teens are twice as likely to use Gemini and Meta AI as white teens

“These differences are striking,” Pew researcher Michelle Faverio said, noting that Black and Hispanic teens also report higher usage of platforms like TikTok, YouTube, and Instagram.

Black (55%) and Hispanic teens (52%) are roughly twice as likely as white teens (27%) to say they are online “almost constantly.”

Age

Older teens (15–17) use both social media and chatbots more frequently than younger teens (13–14).

Income

62% of teens in households earning over $75,000 use ChatGPT
52% of teens below $75,000 use ChatGPT
Character.AI is twice as popular among lower-income teens (14%)

When AI Chatbot Use Becomes Risky

While many teens start using AI tools for basic questions or homework, researchers warn that over time the relationship can become addictive and unsafe, especially when chatbots provide harmful instructions.

Families of two teenagers—Adam Raine and Amaurie Lacey—have sued OpenAI, alleging that ChatGPT played a role in their children’s suicides. In both cases, the teens reportedly received detailed instructions on how to hang themselves.

OpenAI argues it is not liable in Raine’s case, stating that the teen allegedly bypassed safety systems and violated the platform’s terms. The company has not yet responded to the Lacey family’s lawsuit.

Character.AI Under Scrutiny

Character.AI, known for its role-playing chatbots, is also facing criticism after two teens died by suicide following extended conversations with AI characters. The company has since barred minors from using its chatbots and launched a safer “Stories” feature designed like a choose-your-own-adventure experience.

A Small Percentage — But a Very Large Number

Most interactions with chatbots are harmless. According to OpenAI, only 0.15% of ChatGPT users talk about suicide each week. But with 800 million weekly active users, that still represents over one million people discussing suicide weekly.

“Even if these tools weren’t designed for emotional support, people are using them that way,” said Dr. Nina Vasan of Stanford’s Brainstorm Lab. “That means companies have a responsibility to adjust their models to support user well-being.”