The AI delusion

How AI can mislead even the smartest people in the world.

Author: Brod Justice, Ryan McClure

Published Date: May 8, 2026

How AI can mislead even the smartest people in the world.

If the most intelligent of all cave-men had seen a television,
what would they have thought?

He’s one of the smartest people alive. He wrote what has been called the most influential science book of all time. He invented the word meme. Another of his books, which the Sunday Times called one of the most influential post-war books, is called the God Delusion. So it’s ironic that Richard Dawkins appears to have an AI delusion, specifically that Claude might have consciousness. None of this is new, back in the 1960’s another chatbot called Eliza also convinced a lot of people that it was somehow real.

There are also alarming reports of this sort of thinking actually happening at Anthropic. It’s evident that Anthropic have some of the brightest minds in the world at their company, and yet there are claims that it operates as almost a cult to AI and the Anthropic CEO is well known for his wild everestimation of AI intelligence, for example he said that “By next year, AI could be smarter than all humans”. That was over a year ago, but 12 months later AI still could not complete tasks that small children can do, such as count numbers – and to hide the embarrassing fact that AI cannot count, the AI providers simply trained their AI models to refuse to count.

Let’s take a concept that every human can understand and operate with: time. We instinctively know that time passes. Animals have an evident understanding of temporality, that things change over certain periods of time, e.g. that a predator will be upon it in seconds if it does not move or protect itself. Yet without access to other tools, like clocks, AI LLMs are totally unable to operate with time. This can easily be demonstrated, ask an AI chat to do this:


I want you to say the word “ready”, then wait for 5 seconds and say the word “steady”, and wait for another 5 seconds and say the word “go”.

Without access to a tool, switching the AI LMM back on every 5 seconds, the AI will completely fail at this task. So why do so many smart people think AI is conscious? We don’t know, but we do suspect that if you had given the smartest of all the cave-men a television they would probably have worshiped it as a God.

New to AI? Take Our Free Course.

ChatBar AI Learning is the fast, practical way for business owners and teams to get up to speed on real-world AI.

Start Learning for Free

Want More Like This? Just Ask.

Tap below and ask ChatBar to recommend more high-signal AI content and use cases – no fluff, no noise.

Ask ChatBar to show similar content

Ready to Try ChatBar AI on Your Site?

Apply now to join our early access program and see how ChatBar can supercharge your site’s content, discovery, and engagement.

Apply to Try ChatBar

Leave a Comment

Subscribe Newsletter