Sherry Turkle Cautions Teenagers About AI Use
MIT professor and bestselling author Sherry Turkle warned about the dangers of teenagers’ relationships with AI chatbots and AI usage in schools in Latin’s Wrigley Theater on Oct. 28. She discussed these issues and how she addresses them in her newest book in a presentation with Pulitzer Prize and Peabody Award-winning investigative journalist and Latin parent Alison Flowers.
Amid a year of in-class writing assignments to prevent AI usage at Latin and a worldwide teenage loneliness epidemic, Dr. Turkle promoted her recent book “Artificial Intimacy: Who We Become When We Talk to Machines” after a private reception in the Learning Commons.
Dr. Turkle appeared on New York Times columnist Ezra Klein’s podcast in early October, discussing broader ideas like how chatbots are changing human relationships and how they were designed for marketability, not to benefit human relationships. She narrowed her focus during the conversation at Latin, emphasizing the specific dangers of chatbots to teenagers, the most immediate danger being the one-on-one relationship that is created through the chatbot communicating in the first person.
“I argue that when a program goes into the lane of speaking in the first person and saying, ‘I am a person, I am your person,’ it has come into a lane where it shouldn’t be,” Dr. Turkle said during the conversation in the Wrigley Theater. “And no matter how compelling or exciting or titillating that is, one thing leads to another, and you just start to form a relationship with it when actually there is nobody home.”
Among guests at the event were teachers, Latin alumni and parents, some of whom were undoubtedly thinking about how AI would affect them in their personal lives, but many of whom were probably interpreting her warnings from the perspective of an educational institution.
“I think it’s important for parents and adults in general to know that every aspect of your digital technology was deliberately made so that you’re much more aware as a consumer,” Upper School computer science teacher and Department Chair Bobby Oommen said after attending the conversation. “To her point, this is a consumer product.”
Because of the chatbot companies’ objective of marketability, Dr. Turkle argued that users should not fall into the trap of fostering a relationship with the chatbot because that one-on-one design does not make it more effective—it simply makes it more addictive to use.
When you turn away from your chatbot, it doesn’t care if you make a pasta dinner or kill yourself.
— Sherry Turkle
“You have to be mindful that this is an informational tool,” said Latin alum Cecilia Mowatt ‘78, who attended the conversation. “It is not a substitute for a relationship with others. It must work as a relationship with a fact. AI is a tool. It’s an important tool. It can make life easier and faster for you to help you do what you don’t feel like doing sometimes. But it cannot think for you as to who you are, or what you should do for you. You still have that responsibility, and you have to own it.”
Dr. Turkle even argued that chatbots’ use as a tool can be more effective, but it can also prevent the serious harms of developing a relationship with it.
“In the book I’m really careful not to use the word ‘relationship’ to talk about what’s happening between you and the program,” Dr. Turkle said. “I mean, you’re imagining a relationship, and it’s performing a relationship, but there is no relationship. When you turn away from your chatbot, it doesn’t care if you make a pasta dinner or kill yourself. And what kind of relationship is that?”
Dr. Turkle explained that chatbots were designed to speak in this first person conversational way because that design would keep people addicted to using them, arguing that chatbots were created for marketability, not with our best interest in mind.
Mr. Oommen offered support for Dr. Turkle’s argument. “To [Dr. Turkle’s] point of the larger idea of being aware that this is ultimately not something that is in the shape for your best interest, it was made with the bottom line in mind of ‘how do we increase shareholder wealth and shareholder profitability? Let’s design a product that will do that,’” he said. “And so I think as people are more aware, they’ll be much more cognizant of giving it to their kids and how they use it themselves.”
Even for those interested in learning about the effects of using chatbots to help them for work or their personal life, Dr. Turkle’s arguments about the design of these chatbots affecting their outputs helped them understand their personal experiences more.
“I just started using Claude and Gemini three-and-a-half months ago,” said Latin alum Randy Greenberg ‘76, who attended the conversation. “I’m using it for law, for dealing with a couple of lawsuits, and it has saved me so much time and so much money. Is it perfect? No. That’s the problem. The problem is, when I am speaking to these chatbots, I get to the point where I’m typing in, ‘Are you an idiot?’”
Dr. Turkle noted that despite the frustration some feel when using chatbots, many users find solace in AI’s constantly reaffirming manner.
“People were telling me that they loved their chatbot more than their husband, more than their wife, and that they were moving away from their families in order to spend more time with their chatbot, that they were not gonna go to see a human therapist because they had a chatbot therapist who was better,” Dr. Turkle said.
Even though Dr. Turkle saw the effects of people having a close relationship with chatbots, she discussed how she mirrored those exact relationships in an even more dangerous way.
“The most impactful thing that I learned is that even when we are aware of the pitfalls of something, we can fall victim to it, because she had talked about the fact that she used AI to create her own mother and herself as a grandmother,” Ms. Mowatt said.
We’re not as advanced as we think we are.
— Bobby Oommen
Mr. Oommen found the same story to be extremely alarming after realizing how enticing the thought of creating a chatbot for a dead loved one could be, noting, “We’re not as advanced as we think we are. We had mediums and seances and people who did this since the dawn of time. [There were] shamans and ritual people who would speak to the dead on your behalf. It reminded me that we’re still people with these same basic wants and needs and desires, and now we just have a new thing that can exploit that.”
While Dr. Turkle’s argument focused on the vulnerability of the entire population to the alluring idea of AI that can act as a dead loved one, she also discussed her personal experiences teaching college students who use chatbots.
“At stake here really is not just how we write, but how we read, and will reading, as we’ve thought of reading, survive,” Dr. Turkle said. “And I think that educators have to confront this head on and not [make the excuse that] every generation has had a new technology.”
Beyond the possibility of a loss of reading and writing skills at the developmental level because of the use of chatbots in education, Dr. Turkle voiced fears about the use of chatbots for younger people, even suggesting a ban on chatbots for people under the age of 18.
Teenagers in my interviews are the most vulnerable.
— Sherry Turkle
“Teenage years are the most scary years, and [with a chatbot] there is no rejection, there is no anxiety, you’re loved, you’re accepted, it’s all good,” Dr. Turkle said. “[Teenagers] don’t want to be vulnerable. It’s a scary world, and here’s a place where it’s not a scary world and you’re never going to be rejected. You’re always going to feel accepted. The friction of getting to know a new person and of maybe they don’t like you, and actually they may not like you and you’ll have to find another person. Skip it entirely. Teenagers in my interviews are the most vulnerable.”
Dr. Turkle’s points about the future of the use of chatbots among young people reached the appropriate audience of educators, parents and students in the Wrigley Theater, but whether those who hear her calls for change will act on them remains to be seen.
“I think it was super important that [the event] was held in a school to say this is one of the foregrounds where this battle will happen,” Mr. Oommen said. “One of the points she was making is that the train has not left the station. And I think being in a school, it reiterates that point that we haven’t lost this yet. There’s choices everybody in this school can make to counteract some of those forces or issues.”
About the Contributor
Eliza Katz, Investigative Editor
Eliza Katz (‘28) is a junior who is thrilled to continue writing for The Forum for her second year as Investigative Editor. She hopes to reveal and magnify stories that are important to the school community and beyond. Eliza fulfills her passion for marine biology and environmental science through her involvement in the Teen Auxiliary Group at the Shedd Aquarium and the Youth Environmental Press Team. She is a member of the tennis and lacrosse teams, Women in STEM club, and enjoys spending time in nature hiking, camping, and reading.
Comments