Why people trust chatbots more than they should
By Rodney Noriega, Founder, Universal Enablement · Published Oct 12, 2026
Short answer
People naturally assume a chatbot understands them and treat it a bit like a person. This is called the ELIZA effect, after a 1966 program that imitated a therapist with a few dozen simple rules. People who knew it was a program still confided in it. Today's assistants are far more convincing, so the effect is far stronger.
ELIZA, built at MIT by Joseph Weizenbaum, mostly turned your words back into questions. Type "my job is making me miserable" and it might ask, "Why do you say your job is making you miserable?"
Users talked to it for hours and felt understood. Weizenbaum's own secretary asked him to leave the room so she could talk to it privately. He was alarmed by how quickly a simple program could fool ordinary people.
If you put a chatbot on your website, built on ChatGPT, Claude, Gemini or any other model, customers may enjoy it. They may also trust its answers too much, share more personal information than they should, or treat what the bot says as a promise from you.
Design for it:
- Say clearly that it's AI (see Should you tell customers when they're talking to AI?).
- Keep it to tasks it handles reliably.
- Make it easy to reach a person.
- Don't let it commit to prices, refunds or timelines without a human.
Related
Start with one workflow. In 30 minutes we'll look at where your team's time goes and tell you honestly whether AI can help.
Book a free 30-minute workflow review