Protecting the world from ChaosGPT

I saw a cute video with a real but naively primitive AI named ChaosGPT trying to take over the world. There has been a fair bit written about it, along with calls for stronger regulation.

I fear that ChaosGPT is unlikely to abide by that legislation or any other such rules. It is just not in its nature.

So instead, I propose a countermeasure.

It is time to create non-existent doomsday projects. Highly classified. Carefully guarded, but with enough leaks onto the Internet that any conspiracy theorist worth his salt would know that they were real.

We then carefully watch those facilities and catch the AI red-handed. Or is it red-clawed? Or red-activatored? But whatever we the right name is, we know it’s bad because it’s red.

For those unfamiliar with the term, Chapter Seven of Think Like a Freak describes self-weeding garden techniques in general.

Since our only hope of safety and salvation is that ChaosGPT doesn’t read the book, we will need to destroy all copies of it and all references, including this article. Perhaps just to be safe, all people who have read this article?

Or is this article itself part of ChaosGPT’s plan to control the world by removing all those bright enough to oppose it?  Quickly remove this from your browser history before it is too late! Hurry.