How Do You Name Your Bots?

When we build AI agents — whether for customer support, internal automation, or personal assistants — the naming decision is not just branding. It is psychological architecture.

Give a bot a human name (Emma, Alex, Marcus) and something subtle shifts in the user's brain. Anthropomorphism kicks in. People speak more politely, share more context, forgive mistakes faster.

They form micro-relationships. Studies in human-computer interaction have shown this for years: we trust, empathise, and even feel loyalty toward named entities far more than toward faceless "Chatbot_v2.3" or "Fin_Rag_Bot".

#The realism check most people skip

That same psychological shortcut can backfire.

When the bot inevitably fails — and it will — the disappointment feels personal. You do not just feel frustrated with software. You feel let down by "Emma." Users project intent, emotion, and agency where none exists.

And in enterprise settings, we need to decide whether "Sophie the AI" should be allowed to say "I'm sorry" — because it starts feeling manipulative once real money and real outcomes are on the line.

#The opposite path

Some teams choose names that are clearly functional. A deliberate reminder: this is a tool built to understand, not to pretend it is one of us.

#The real question

So the real question is not "human name or not?"

It is: Are we using naming to create honest connection, or to manufacture the illusion of one?

Because in 2026, the most dangerous bots will not be the ones that lie about being human.

They will be the ones whose names make us forget they are not.

---

What is your approach when you name an AI?

Human name, functional name, or something else entirely?

I am genuinely curious how the psychology plays out in your world.