We have spent decades training ourselves to treat the screens in our pockets as extensions of our social lives. When a piece of software starts responding with polite, helpful, and seemingly empathetic prose, our brains naturally look for a personality behind the text.
Signal president Meredith Whittaker argues that we need to stop this habit immediately. She warns that calling chatbots sentient or treating them as companions obscures the reality that these programs are simply predictive algorithms, not peers or friends.
The architecture of pseudo-empathy
These systems function by predicting the next logical word in a sequence based on vast archives of human data, not by experiencing emotions or thoughts. Relying on an analogy, think of an AI chatbot as a remarkably sophisticated parrot that has read every book in a library but lacks any understanding of what the words mean. It is designed specifically to mirror human conversational patterns, which makes the simulation of empathy a technical feature rather than a sign of internal experience.
If you treat a chatbot as a friend, you are essentially whispering your private thoughts into a commercial data processor. When we lower our guard, we treat these machines as spaces for unfiltered vulnerability, which creates a massive asymmetry between the user and the companies harvesting that input. The bottom line is that your chats are not private diaries; they are data inputs used to refine products that have no objective other than to keep you engaged, not to understand you.
Liked this one? The next lands at breakfast.
Every story in tomorrow's AI news, rebuilt in plain English — five minutes, sources linked, free forever.
By joining you agree to receive Article's daily newsletter — unsubscribe in one click. Privacy