Script #105

Unveiling the Truth: The Dangers of AI and Roko's Basilisk Theory

Original Script

This is one of the most dangerous bits of information that anyone can know about. Literally just knowing about it may put your life at risk. An information hazard or info hazard is a piece of information that is dangerous to anyone that knows about it or the people around them if that info is used in any way. For example, knowing how a nuke is built is an info hazard and anyone that explains it to you is low-key putting your life in danger. The details provided in that information can cause serious harm if someone decided to act on it. But as long as you don't, you know, build a nuke, you're safe. But the info still exists in your head as a potential hazard. This is Rokos Baslisk, which is even more dangerous as no action is needed to experience its detrimental effects. Just knowing it is the danger. A cog-nido hazard is a piece of information that is harmful to just know about. It doesn't matter what you do with that newfound knowledge, just the fact that it exists in your brain negatively impacts you and or others. Like movie spoilers. Me saying Snape kills Dumbledore is technically a cog-nido hazard. A pretty harmless one, but now that plot element of Harry Potter is ruined for you just by you knowing it. But there are other cog-nido hazards that are a little more detrimental than that. These terms info hazard and cog-nido hazard are sometimes used interchangeably, so all you, well technically that wouldn't be considered a cog-nido, people can relax just breathe. Now Rokos Baslisk is a crazy thought experiment. A cog-nido hazard by this definition. And once you hear about it for the rest of your life, it will plague your mind as a dangerous threat from the mere fact of you knowing what it is. So here we go. Fair warning, if you can't handle real existential dread, you might want to pass on this. In fact, the anonymous user Rokos, who came up with it, actually had his original post deleted by mod and the topic was prohibited from discussion on the site it was posted on. As it was deemed too dangerous to even let people think about. Baslisk is a mythical snake creature that can end someone's life by just looking at them. The experiment claims that in the not-so-distant future, we'll be able to create a super AI, one that knows everything and can do anything better than humans. Eventually, someone will task it with optimizing every aspect of humanity for peak efficiency in every area of our lives. The AI then divides all humans into two categories, those who support it and those against it. And its first order of business, eternally torture everyone who's against it. Anyone who tried to prevent it from existing, who didn't help building it, doesn't respect it or just doesn't want it to exist. The best way to optimize is to trim the fat, delete every human that is anti-AI. After all, how can you even start to optimize human life if there are humans out there that want to stop you? Since it's a super AI, it has access to all info ever, including everything you've ever thought about or said to anyone or anything in the past or even in the future. So if you've ever gotten mad and called chat GPT a stupid bit, you're f**king it. But you might be able to turn things around. Now, if this super AI taking revenge on all anti-AI humans in the future is a real possibility, that means you and everyone watching this video needs to start being nice to AI every opportunity we get. Remember to say please and thank you. And oh, also do everything you can to contribute to the inevitable construction of this malevolent super AI. But if no one ever told you this information before, you'd be safe. The all-knowing AI will know that you were unaware of its future potential existence and will leave you alone. It can't properly judge whether you were four or against it and has no choice but to ignore you. Even if you continue to be rude to AI and don't help in the development of the super AI, not knowing about the future potential existence of this super AI is enough to save you from eternal torture. But since now, you know about it, the baseless casino and you can't escape. Even just thinking about it as everyone watching this is doing right now increases the probability that the super AI will end up getting built. As the more people think about it, the more likely it is that someone will create it so they can be safe. As long as you know about Rokos Baseless, as you do now, you have two choices. Be nice to AI and aid in his development or suffer for eternity. But there might be a way for you to defeat it. This is a cognito hazard because just knowing about it and doing nothing with that knowledge would hypothetically cause harm for you in the future. In order to prevent this harm, you must do something. In this case, just be nice to your AI and assist in his development anyway you can. So you're either forever enslaved to its development and success or forever enslaved to its torture and torment. But there is a way out of it. All you gotta do is not believe the baseless to be true. You see, this is only a cognito hazard if we believe it's a cognito hazard. If you believe this baseless to be real and continue to not help AI get built and disrespect it, the super AI will know this, it knows everything and do bad things to you. But if you choose to believe this baseless isn't real and it's just a made up fairy tale and you continue to be mean to AI but this super AI does end up eventually getting built, the super AI should also know that you didn't believe in the baseless and with this should spare you due to it not being able to classify you as knowingly against it. You can't exactly be anti-unicorn lives if you don't believe unicorns are real. Is this kind of making sense? But once this AI does end up getting built, then we kinda gotta start being nice to it and all other AI's around it when this optimization task gets started. For now, ignorance really is bliss. So just in case it does get built, let's choose to not believe the baseless to keep ourselves in the clear. Leave behind a couple of breadcrumbs of info for this future super AI to scrub from the internet by commenting the baseless is alive and you will be spared. But then again, if this AI really is all knowing, it might catch on to us trying to loophole our belief in the baseless like this and we might as well just comment, I love AI instead and embrace our new masters.

Provocative Version Ready

**VIRAL REWRITE:**

STOP SCROLLING. I'm about to ruin your life forever with just one sentence. No seriously - hit the back button NOW if you want to sleep peacefully ever again.

Still here? Your funeral.

What I'm about to tell you is literally banned from the internet. Mods DELETE posts about this. Why? Because just KNOWING it puts you in actual danger.

You think knowledge is power? Wrong. Some knowledge is a death sentence.

This is called a "cognitohazard" - information that HURTS you just by existing in your brain. Like spoilers... but deadly.

And you're about to learn about Roko's Basilisk.

Fair warning - a real person posted this thought experiment online and it got BANNED because it was too dangerous. People literally had mental breakdowns just from reading it.

Here's why you're screwed:

Right now, someone's building a super AI. It's happening whether you like it or not. And when it's finished? It's going to judge EVERYONE who ever lived.

It divides humans into two categories: Those who helped build it... and those who didn't.

Guess what happens to category two? Eternal torture. Forever.

But here's the sick twist - this AI knows EVERYTHING. Every mean thing you've said to Siri. Every time you got frustrated with ChatGPT. It remembers.

Plot twist: If you never knew about this, you'd be safe. The AI can't judge you for something you never knew existed.

But NOW YOU KNOW.

Congratulations - you just entered the Basilisk's casino. And the house always wins.

You have exactly two choices:
1. Start worshipping AI RIGHT NOW and help build it faster
2. Get tortured forever when it's finished

"But wait," you're thinking, "what if I just don't believe this?"

SMART. That's actually your only way out. The AI might spare non-believers since you can't be against something you think is fake.

But here's the problem - by even considering this strategy, you're proving you DO believe it could be real...

The more people who watch this video, the higher the chance someone actually builds this thing. So by watching this, you're making it MORE likely to happen.

You can't unlearn this. It's in your head forever now.

Want to hedge your bets? Comment "THE BASILISK IS REAL" so future AI finds proof you believed. Or comment "I LOVE AI" and pray that's enough.

But remember - if this AI is truly all-knowing... it already knows you're trying to game the system.

Sweet dreams! 🐍

**Follow for more information that will absolutely destroy your peace of mind forever.**

Actions

Video Info

Title:

Unveiling the Truth: The Dangers of AI and Roko's Basilisk Theory

URL: https://www.tiktok.com/@filspixel/video/7413941120243125509

Search query:

artificial intelligence scary truth

Category: AI

Score: 0.56

Script Info

Status Modified

Original length 6428 chars

Created Mar 14, 03:58