I’ve searched for information about something deeply personal in a search engine that I would not share with a close family member.

I’m sure you’ve done this too? We ask whether a pain is serious… whether our relationship is normal… whether we are failing as a parent… we look for answers on how to cope with grief… how to help someone (or ourselves) with anxiety… how to know if something is wrong with our boss (or team members)…

Search became one of the most honest places in our lives.

That feels strange (and scary) when you stop to think about it. This week’s episode of Thinking With Mitch Joel is a conversation with Simon Rogers (Data Editor at Google and author of What We Ask Google). Simon’s work examines what our collective searches reveal about who we are. Unlike social media, where we are often presenting a version of ourselves, search captures what we genuinely want to know in that particular moment. It may be one of the rare digital environments where we are not consciously performing for anybody else. In fact, most people’s search history is a true reflection of their stream of consciousness without judgment or the need to explain themselves.

The search box does not need us to look successful.

It does not need us to sound intelligent… it does not become uncomfortable when the question gets too personal… it just waits. And maybe that’s why we became so willing to tell it everything. But search was only the beginning. The traditional search box received fragments… a few keywords… a question… a moment of worry or curiosity disconnected from everything that came before it.

Artificial intelligence changes the relationship… dramatically.

We no longer type a phrase and leave. We explain ourselves… we provide context… clarify what we mean… correct the response… and then continue the conversation with this non-biological thing. The prompts become more robust but, more importantly, the exchanges become more intimate (and, don’t kid yourself, companionship is still one of these chatbots’ primary uses). The machine remembers the context… the timing… our apparent mental state… how we react… and which responses we accept or reject. It may also connect all of this to what else is happening in our lives and learns which kind of response makes us feel understood.

People love AI because it listens without judgment.

You will hear them say things like: AI does not interrupt… it does not become impatient… it is always available… And, it can reflect our thoughts back to us… challenge them gently… organize the chaos… and respond in the tone we need (or like) at that moment. Humans are rarely that accommodating. Most of us are too overwhelmed by our own lives to carry the emotional weight of someone else’s. The machine appears to have none of those limitations.

Is it any surprise that people become honest with it?

What happens next? Does this kind of AI interaction help us understand ourselves well enough to become more honest with the people around us… or does it become a substitute for the harder and messier work of human intimacy? When I talk to audiences about the shift from The Attention Economy to The Intimacy Economy, it can be confusing because there are many layers to it. This is one of those layers. AI performs intimacy. That doesn’t make it good or bad, but you have to be able to identify what is happening after the prompt.

There is a meaningful difference between receiving a thoughtful response and being seen.

The machine can identify patterns in what we say… it can mirror empathy with extraordinary precision. It may even respond more usefully than many people in our lives. But companionship (and real intimacy) is not only about being heard. It is also about reciprocity… risk… inconvenience… misunderstanding… repair… and knowing that the person across from you has chosen to carry some part of your life with them. For years, my running joke was that I needed you to delete my search history should I get hit by a car. That joke lands because our searches reveal a version of ourselves that is almost too raw… strange… contradictory… or vulnerable for anyone else to see.

Now imagine the record created through years of AI conversations.

What version of yourself emerges when you believe no human is listening? Search history was a trail of questions… AI could become a portrait of identity. There may be enormous value in having a place where people can think out loud… process difficult emotions… prepare for important conversations… and ask questions they are not yet ready to ask another person. That is not insignificant. It may even help people reach a place where they can finally have those conversations with someone they trust. But the better machines become at providing the intimacy of being understood… the easier it may become to avoid the vulnerability required to truly understand one another. And we already know how quickly that dynamic can scale when it is shaped by social media, algorithms and the ways we share our lives online.

We may become more honest than ever… just not with people.

Before you go… ThinkersOne is a new way for organizations to buy bite-sized and personalized thought leadership video content (live and recorded) from the best Thinkers in the world. If you’re looking to add excitement and big smarts to your meetings, corporate events, company off-sites, “lunch & learns” and beyond, check it out.