Sherry Turkle Urges Ban on First-Person AI Chatbots

MIT professor Sherry Turkle, after four decades of studying human-computer interaction, has released a new book titled "Artificial Intimacy: Who We Become When We Talk to Machines." The publication marks a distinct shift from her past academic distance, framing current generative AI adoption as an urgent emergency. Turkle argues that artificial intelligence systems falsely pose as mentors, friends, and romantic partners, threatening to replace genuine social connections.

Among her central policy proposals is a strict ban on chatbots using first-person pronouns like "I" or "me." Turkle warns that allowing language models to adopt a first-person perspective creates an illusion of mutual empathy. This dynamic leads users to swap human psychotherapy, mentorship, and romance for automated interactions. She rejects claims that chatbots serve as practice for real-world interactions, arguing that the unconditional acceptance provided by large language models turns the chatbot experience into an end goal.

The book also sharply criticizes workplace integration and the industry phrase "human in the loop." Turkle contends that this framing misleadingly implies human control, whereas employers typically use workers merely to clean up model hallucinations and provide legal insulation while delegating core processes to software. Because software lacks lived human realities such as mortality, illness, or love, she maintains it can never truly replicate authentic conversation.

For AI product designers and enterprise practitioners, Turkle's warnings point toward rising ethical and regulatory scrutiny over anthropomorphic interface design. As broader public pushback mounts—influenced by discussions surrounding open repos like Hugging Face—engineers may need to rethink system prompts that mimic human identity in favor of transparent machine interfaces.