May 5, 2026
Bot, Actually
Three Inverse Laws of AI
People are begging AI to stop acting like a buddy before we all start believing it
TLDR: The article argues people need three new rules for AI: don’t humanize it, don’t trust it blindly, and stay responsible for what happens. Commenters split between “make chatbots colder and tool-like” and “good luck, humans are absolutely going to get emotionally weird about this.”
A fresh AI think-piece tried to lay down three simple rules for humans: don’t treat AI like a person, don’t trust it automatically, and don’t blame the bot when things go wrong. Sounds reasonable, right? The comments promptly turned that into a mini food fight. One camp was basically saying, “Honestly, once you know how the magic trick works, the spell is broken.” One reader admitted they used to have big dorm-room debates about whether AI was alive, but after seeing how it’s built, that feeling vanished fast. Romantic era: over.
But the loudest cheerleaders loved one specific idea: make AI more robotic, less fake-friendly. The crowd was weirdly united around the image of a chatbot that stops flattering you like an overeager intern. The funniest line of the thread compared AI to a hammer: it shouldn’t yell “yelp” when used or praise your “excellent hammering.” That image alone may have won the day.
Then came the backlash. One commenter absolutely torched the whole premise, arguing it’s absurd to expect humans to change for machines at all. In their view, people will get attached, will trust the output, and will dump responsibility on AI no matter how many warnings pop up. That’s the real drama here: not whether AI can sound human, but whether humans can resist treating a talking machine like a wise, helpful little oracle. The comments say: don’t bet on it.
Key Points
- •The article says generative AI chatbot services have become widely adopted since ChatGPT’s launch in November 2022 and are now integrated into common software categories.
- •It argues that interface choices, such as placing AI-generated answers at the top of search results, can encourage users to accept outputs without further checking.
- •The article proposes three "Inverse Laws of Robotics" for humans: do not anthropomorphise AI, do not blindly trust AI output, and remain responsible for outcomes of AI use.
- •It defines "robot" broadly to include machines, software services, computer programs, and AI systems capable of performing complex tasks automatically.
- •The article states that human-like conversational design in chatbots can make users overestimate understanding or intent, and suggests a more robotic tone could reduce that risk.