I watched Gary North on some late-90s TV interview once. He was dead serious, staring into the camera: “Set your computer’s clock to January 1st, 2000. I dare you. It’ll stop working.” I remember thinking: that’s not how any of this works. Change the date and reboot, worst case you get a Y2K bug in whatever software reads it wrong, and from my testing there were only two on a PC that had an error. One of those was Excel where the programmers were lazy. The used decimal for the date instead of binary coded decimal. The computer itself doesn’t care. It just tracks and displays the time. North didn’t know that. And he was the guy millions of people were paying to tell them how to survive the collapse of civilisation. He wasn’t a crank on the internet fringe. He ran a newsletter for over two decades. He called Y2K “the biggest problem the modern world has ever faced.” He moved his family to rural Arkansas. He told people to stock food, gold, guns. He was the loudest voice in a very crowded room of people convinced January 1, 2000 was the end of the modern world. It wasn’t. Nothing happened. Planes didn’t fall. Banks didn’t collapse. The lights stayed on. Here’s the part that’s actually interesting, though. Bill Bonner made an argument at the time that I think about a lot: the reason nothing happened might be because of guys like North. Companies spent something like $300 billion worldwide fixing the two-digit-year bug before the deadline hit. All that panic, all that fear-mongering, might have been the thing that prevented the disaster it was warning about. So which was it? Was North a fraud who got lucky, or an alarmist who accidentally saved everyone by being loud about a real (if wildly overstated) problem? I don’t think it’s either. I think he was a guy who didn’t understand the technology, selling total confidence about it anyway. That’s the part that stuck with me watching that clip, not that he was wrong, but that he was so sure, about something he plainly hadn’t looked into past the headline. I see the same move constantly right now. People with zero hands-on time in a tool selling total certainty about what it can and can’t do. Confidence isn’t the same thing as knowing. If you want the version of this that’s actually useful, prompts built from real use, not confident guessing, that’s what the AI Prompt Guide is for: https://link.ckv.to/prompt-guide Regards, |
