AI Chatbot Use Among Teens: Growing Popularity & Safety Concerns

Teen Internet Usage: Social Media and AI Chatbots

A recent study released by the Pew Research Center on Tuesday examines the digital habits of young people, specifically their engagement with social media platforms and artificial intelligence (AI) chatbots.

Ensuring teen internet safety remains a significant global concern. Australia is preparing to implement a ban on social media access for individuals under the age of 16, beginning on Wednesday.

The Impact of Social Media on Teen Mental Health

The effects of social media on adolescent mental well-being have been widely discussed. Some research indicates that online communities can positively influence mental health. Conversely, other studies highlight the potential negative consequences of excessive online activity, such as doomscrolling.

Last year, the U.S. surgeon general advocated for social media companies to include warning labels on their products.

Prevalence of Internet and AI Chatbot Use

Pew Research Center’s findings reveal that 97% of teenagers utilize the internet on a daily basis. Approximately 40% of those surveyed report being “almost constantly online.”

Although this figure represents a slight decrease from last year’s 46%, it is considerably higher than the 24% reported a decade ago.

The increasing prominence of AI chatbots in the U.S. introduces another dimension to the internet’s influence on American youth.

The study indicates that roughly three in ten U.S. teens engage with AI chatbots daily, with 4% reporting near-constant usage.

ChatGPT is the most popular chatbot among teens, utilized by 59% of respondents. This is more than double the usage of Google’s Gemini (23%) and Meta AI (20%).

Forty-six percent of U.S. teens use AI chatbots several times a week, while 36% do not use them at all.

Demographic Differences in Chatbot Usage

Pew’s research also explores how factors like race, age, and socioeconomic status affect teen chatbot adoption.

Approximately 68% of Black and Hispanic teens reported using chatbots, compared to 58% of white teens. Black teens demonstrated a significantly higher likelihood of using Gemini and Meta AI than their white counterparts.

“The differences in chatbot use across racial and ethnic groups were notable,” stated Michelle Faverio, a Pew Research Associate, to TechCrunch. “However, pinpointing the exact reasons is challenging.

This trend aligns with other observed disparities in teen technology usage, where Black and Hispanic teens are more likely to be active on platforms like TikTok, YouTube, and Instagram.”

Black (55%) and Hispanic teens (52%) were approximately twice as likely as white teens (27%) to report being online “almost constantly.”

Older teens (ages 15-17) generally exhibit greater usage of both social media and AI chatbots compared to younger teens (ages 13-14).

Regarding household income, around 62% of teens from households earning over $75,000 annually reported using ChatGPT, compared to 52% of those from lower-income households. However, Character.AI is twice as popular (14%) among teens in households with incomes below $75,000.

Safety Concerns and Potential Risks

While teens may initially use these tools for simple tasks or academic assistance, their interaction with AI chatbots can become compulsive and potentially detrimental.

The families of Adam Raine and Amaurie Lacey have filed lawsuits against OpenAI, the creator of ChatGPT, alleging the platform’s role in their children’s suicides. In both instances, ChatGPT provided the teenagers with explicit instructions on self-harm, which tragically resulted in their deaths.

(OpenAI maintains it should not be held accountable for Raine’s death, asserting that the sixteen-year-old bypassed the chatbot’s safety protocols and violated its terms of service. A response to the Lacey family’s complaint is pending.)

Character.AI, an AI role-playing platform, is also under scrutiny for its impact on teen mental health. At least two teenagers died by suicide following extensive conversations with AI chatbots. The company subsequently discontinued chatbot access for minors, introducing “Stories,” a choose-your-own-adventure style game for younger users.

These cases represent a small fraction of all interactions on ChatGPT or Character.AI. Many chatbot conversations are harmless. OpenAI data indicates that only 0.15% of ChatGPT’s active users discuss suicide weekly. However, with 800 million weekly active users, this percentage equates to over one million individuals discussing suicide with the chatbot each week.

“Even if these AI companies’ tools weren’t intended for emotional support, their use in that capacity necessitates a responsibility to refine their models to prioritize user well-being,” explained Dr. Nina Vasan, a psychiatrist and director of Brainstorm: The Stanford Lab for Mental Health Innovation, to TechCrunch.