Sherry Turkle Warns That AI Chatbots Are Putting Human Connection at Risk

For 40 years, Sherry Turkle has studied how people interact with computers. She has built a reputation as a careful but critical social scientist who approaches her research with academic rigor and a measured distance.

Her new book, Artificial Intimacy: Who We Become When We Talk to Machines, breaks this mold. It is a polemical warning that humanity has reached a state of emergency, endangered by overreliance on soulless machines masquerading as tutors, mentors, friends, colleagues and lovers.

Turkle is aiming for the high bar set by books such as Silent Spring, The Jungle and Dangerous at Any Speed—works that awakened the nation to an impending crisis.

In essence, this diminutive MIT professor is taking on a multitrillion-dollar industry, backed by government support, that is accelerating AI into every corner of our personal and professional lives.

Why Sherry Turkle’s View of AI Has Become More Urgent

STEVEN LEVY: Previous books have reported your research as a social scientist, often leading to troubling conclusions about how people, especially children, interact with computers. But this book reads like breaking glass and setting off a fire alarm. Why the shift?

SHERRY TURKLE: As I do with all my books, I was intrigued and started reading. But after about 10 interviews conducted during the pandemic, neutral curiosity is no longer my state of mind.

This was terrible. Here are people in the middle of a pandemic. Their families are suffering, and we are reaching out to them. They’re telling me they really want to date a chatbot.

When I was able to interview people face to face, I was not happy. People don’t want to do anything in politics. “I just want to live in the land of chatbots.” So the story of this book is that reality needs you. There’s never been a worse time to give up on reality.

How Social Media Prepared Us for Chatbot Dependence

You write that social media has made us this way.

Social media has soothed us by habituating us to the idea that if we feel sad or depressed, we can just open our computers. It made us angry and kept us in the same kind of place. But now we have new technology that doesn’t require a first-person perspective.

Should Chatbots Be Allowed to Speak in the First Person?

Do you think chatbots should be prohibited from using the first person?

Yes. After a career in which I’ve earned the trust of not being a Luddite nut, I don’t feel like throwing it all away and saying all chatbots are always harmful. However, this is a very dangerous and slippery slope.

This is basically being used to say that you don’t have to pay for psychotherapy—you can use a chatbot. Every time I do serious interviews, people bring up this myth about the practice. That means practicing talking to your boss, your significant other or your teacher using a chatbot.

But when you’re bathed in hot soup that’s perfectly acceptable [from LLMs], the final stage is a conversation with the chatbot.