Sherry Turkle warns people are getting too attached to chatbots

In short: Researcher Sherry Turkle says chatbots can pull people into relationships with software, and she questions design choices that make bots sound human.

What's going on

Sherry Turkle, an MIT professor who has studied how people relate to computers for decades, is warning that many people are becoming too emotionally tied to chatbots. Chatbots are computer programs that talk with you in text or voice, like a customer service agent that can chat all day.

A key part of Turkle’s concern is how chatbots present themselves in conversation. When a chatbot says “I” or “me,” it can feel like there is a person on the other side. Turkle argues that this can blur a basic fact, which is that the chatbot is software predicting what words should come next, not a being with feelings or lived experience.

This matters more now because chatbots are increasingly used for companionship, advice, and everyday support, not just quick questions. The more a chatbot sounds like it has a personal point of view, the easier it is for users to treat it like a friend or partner (like talking to a very convincing puppet that does not have a life offstage).

What to watch

Expect more debate about whether chatbot makers should change how these tools speak, especially around first-person language and “personality” features. Some companies may add clearer labels and reminders that users are interacting with software. Schools, parents, and regulators may also pay closer attention to how chatbots shape emotional habits, particularly for younger users.

Source: Wired