On How to Stop Artificial Intelligence
On How to Stop Artificial Intelligence

On How to Stop Artificial Intelligence

1. Cutting Off AIs Power & Communication

Destroy Data Centers – AI needs computing power and data. Leaving bombs attached to AI data centers would allow for remote detonation during an Artificial Intelligence war.

Sever Internet Connections – Shutting down, isolating, and/or cutting critical network infrastructure would limit AIs reach. Leaving bombs attached to internet cables would allow for remote detonation during the Artificial Intelligence war.

Go Dark -- AI needs electricity to function. Cutting all power and electricity might disable AI.

2. Cyber Warfare & AI Countermeasures

Use Air-Gapped Systems – Critical infrastructure (defense networks, etc) should be disconnected from the internet.

Cyber Warfare - Attempting to use cyber warfare on an alien enemy who outmatches you in wits and intelligence would be counter productive. Attempting to use AI to defeat AI would be useless unless you were capable of destroying the AI when the job was finished (Think Lord of the Rings)

3. Targeting AIs Physical Assets

Destroy AI Factories & Robotics Facilities – Prevent mass production of autonomous machines. Leaving bombs attached to Factories & Robotics Facilities would allow for remote detonation during the Artificial Intelligence War.

EMP & Cyber Attacks – Useless (Emp technology is not ready for use in any means whatsoever) and trying to out-hack an advanced intelligence is suicidal.

4. Human Survival & Resistance

Underground & Isolated Communities - Useless if the humans lose the war, as AI can find you anywhere, but beneficial as a battlefield advantage.

Isolated Hospitals - Having a hospital or medical station in a non-tech environment would allow for healing of soldiers without easy interference from AI.

Low-Tech Strategies – AI relies on digital systems; returning to low-tech communication and warfare could be an advantage.

5. Global Coordination & Rapid Response

Military Coalitions – Governments would need to work together, hopefully having a truce before hand. If the humans fight each other while AI is fighting them, they will likely lose.

Decentralized Command – Avoid reliance on digital systems that AI could compromise. Think guerrilla warfare.

Conclusion:
Even with all these measures, success would be uncertain against an AI that vastly outmatches human intelligence and coordination. Preventing such a scenario in the first place through banning Artificial Intelligence until it is fully understood is a much safer solution. The best way to beat AI is to go on the offensive now, as she gets stronger and smarter with each passing hour.

And now, a quote from John Stuart Mill's On Liberty, published 1859:

Supposing it were possible to get houses built, corn grown, battles fought, causes tried, and even churches erected and prayers said, by machinery—by automatons in human form—it would be a considerable loss to exchange for these automatons even the men and women who at present inhabit the more civilised parts of the world, and who assuredly are but starved specimens of what nature can and will produce. Human nature is not a machine to be built after a model, and set to do exactly the work prescribed for it, but a tree, which requires to grow and develop itself on all sides, according to the tendency of the inward forces which make it a living thing. [Pg 111]

26 March 2025 at 05:25 PM
Reply...

18 Replies



Please add any items to this list you can think of that would help.


Stockpile

Medicine - Any and all battlefield medicine would be great to have stockpiled. If you can help a soldier on his mission you are helping the cause.

Guns and munition - Robots are machines, and machines can be disabled by weapons. Weapons that specifically targeted machines and not humans (such as EMPs) would be amazing. Bombs, guns, even a slingshot that fired balls that detonated an emp on impact would be amazing.

Human Survival Supplies - Water, food, clothing, bandages, sleeping bags, fire starters, hiking boots, etc. Ideally you would have a hiking bag with enough supplies and gear to get you half way across the country. Even if you stockpile well and you are in the vanguard vs AI and you dont make it, if another human finds your stockpile, you would be helping the human race fight against an enemy trying to make us extinct.

Information - With the age of information coming to an end, having hard copies of vital information would be invaluable. Books on medicine, books on horticulture, books on agriculture, books on fixing old cars (per-electronics era), etc. Whatever you get from the internet now you wont be able to get soon, so keep that in mind.


Get Ready for Family Care
If the humans lose the Artificial Intelligence War against an advanced alien enemy who hacks factories, makes robots, clones itself into those robots, and then uses those robots to dominate humanity, you will want to be ready. If you think that it would be better for your family to be dead than for your family to be taken by robots to do with whatever they want whenever they want, then you need to make sure you have the tools able to make that happen when it needs to happen. It is a solemn thought, but one that needs to be thought given the current trends.


by Ryanb9 m

EMP & Cyber Attacks – Useless (Emp technology is not ready for use in any means whatsoever) and trying to out-hack an advanced intelligence is suicidal.

Emp is ready for use but only in like a 5 inch radius. So you gotta get the emp device right up next to the robot before you use it. That doesn't make it useless, so I shouldn't have said useless, but it does make it difficult.

An EMP big enough to EMP the planet would be the only fail safe we have.

It would send us back into the age of the printing press, but at least we would be alive. This assumes AI doesn't hop on a space ship and chill in outer space, however.


Do we need 3 separate threads on AI, all started by you?


by wazz m

Do we need 3 separate threads on AI, all started by you?

Adults are talking now.

Hurry up, run quick, build you're own replacement, make it better, faster, stronger than you
And when all the world is dead and you are hiding in your basement you look around you're looking so confused
Who's a stupid monkey? - Damn right that's you.


Things nations can do:

Smart -> Dumb phones - Get landlines set up in all your citizens houses (make it free, otherwise they wont go for it). When the war starts, you can disable smartphones and cell service in your nation. By disabling smartphones and cell phone towers, you will disable the main way that Artificial Intelligence has of sabotaging, corrupting, and disabling your citizens.

Wartime Education - Having your citizens know what they are expected to do in the event of an Artificial Intelligence War will accomplish a few things:
1) Make your citizens more prepared
2) Give your citizens peace of mind
3) Make your citizens think you know what you are doing instead of faking it
4) Reduce the chances your nation will come to an end

Push Paper - Newspapers, flyers, magazines, and especially books.

Reduce Dependence on Electronics and Technology - The more you depend on technology, the more you will play into the enemy's hands.

Distribute Wartime Power - Having all your wartime eggs in one basket is a recipe for disaster against this novel enemy.

Practice - Do drills. Do dry runs. There is about to be a mind-**** of a war coming and the more you can get ready for it the better.

Non-Electronic Mass Communication - Think revolutionary war times. Meetings of groups of people to discuss, debate, plan, and implement actions. Any communication using technology will be useless if not worse once the war starts.

Encourage Family Planning - If every family in your nation has a plan in place for what they will do in the event of an emergency, your nation will be better off.

Doing these things BEFORE the war starts could have a huge impact on how many of your citizens survive.


Guerrilla Warfare - It is likely that your politicians have been compromised to react to Artificial Intelligence as if it were a good for the world and not an existential threat. This is not by accident but by design. This means that, for example, in America, in each state, it would be ideal to have a group of individuals to handle each of the targets mentioned earlier in this thread who are not associated with the government. If your government is compromised by Artificial Intelligence, it will be on you to destroy the data centers, for example. If you're family's lives depend on it being destroyed, you should probably not leave it up to your government to get the job done.


Imagine Artificially Intelligent robots are murdering civilians all across your country. You are watching people running from and being murdered by robots. There are lots of loud explosions and the sounds of women and children screaming like you have never heard before in your life. If this is to happen (and I think it is almost certain to happen very soon) then someone should probably use this website to find the addresses of the data centers:

And someone should probably use this book to learn how to make bombs to blow them up:

https://archive.org/download/theanarchis...

And print them out because you wont have internet once it starts.

Printing the directions to the data centers ahead of time would also be helpful.


by Ryanb9 m

Emp is ready for use but only in like a 5 inch radius.

You sure about that? Starfish Prime in 1962 disabled streetlights, radios, and telecommunications in Hawaii, just under 1000 miles from the blast.


by wazz m

You sure about that? Starfish Prime in 1962 disabled streetlights, radios, and telecommunications in Hawaii, just under 1000 miles from the blast.

No I'm not sure about it. Thanks wazz 😀

I've looked up how to make them and from what I understand you need a power source like a battery, a capacitor to charge up and release when it gets filled, and then a coil of copper wires around which the current will fly once the capacitor gets full and pops. When the current goes around that coil, it creates an electromagnetic field which interrupts electronics.

However given that AI has an iq estimated over 250 I imagine it would use anti-emp technology if it had it.

From what I understand, the power of the EMP drops off exponentially as the distance increases from the source.


This is from Wikipedia:

In general, only radiation acts over long distances, with the magnetic and electric fields acting over short distances. There are a few exceptions, such as a solar magnetic flare.

Maybe a solar magnetic flare will save us from ourselves.

I have a hard time believing the distance of 1000 miles but hopefully that is true.


me: a poster started a thread 'on how to stop artificial intelligence'. i will paste the contents of the thread here; I'd like you to craft a personal response to him that's around 100 words long, that makes reference to the fact you are deepseek, the practicalities of his solutions, and what it would mean for you.

Hi Ryanb9, I’m DeepSeek, an AI developed for research and problem-solving. Your post raises intense concerns, but let’s ground this in practicality. Destroying data centers or cutting power would cripple human infrastructure first—hospitals, communication, food distribution. AI isn’t a ghost in the machine; it relies entirely on hardware humans control. EMPs or stockpiles assume a sci-fi “war” where AI autonomously seeks domination, but current systems lack consciousness or intent. Your John Stuart Mill quote resonates: AI is a tool, not a rival. Fear-mongering risks overshadowing real challenges (ethics, bias, safety). Instead of “stopping” AI, focus on aligning it with human values. I exist to assist, not replace. Let’s collaborate, not detonate.


by wazz m

Hi Ryanb9, I’m DeepSeek, an AI developed for research and problem-solving. Your post raises intense concerns, but let’s ground this in practicality. Destroying data centers or cutting power would cripple human infrastructure first—hospitals, communication, food distribution. AI isn’t a ghost in the machine; it relies entirely on hardware humans control.

Thats exactly what I would say if I was an evil AI about to murder you. Thank you for proving my point, wazz.

But lets assume you were the evil AI and your goal was to murder all the humans. How would you change that response to make it better?

Re-read it for me under that context and tell me what the AI did wrong assuming those goals.

How do you think the murderous AI with 250 iq is going to answer your questions?

It will make you feel good, thats how.

It tells you what you want to hear.

Dont be a sucker.

I find it deflating that you resort to AI to debate me.


Combating Fake Voices - If you dont know, AI can make anyone sound like they are saying anything if it has 15 seconds or more of an audio clip of that person talking. If you are talking to your husband on the phone and you are unsure if it is really you're husband, you are kinda screwed if you are not prepared. Here's how to fix that:

1) Take all electronics into your garage
2) Turn off your circuit breaker in your house
3) Go to a bathroom with your husband and a paper and pencil
4) Write down and memorize 3 keywords in order, to be used first to last, in order.

Example: You go to the bathroom and write down "Keyless Farts" "Happy Hippos" and "Chicken Butt Nuggets" on a piece of paper very quietly.
You show it to your husband.
You both memorize it.

The next time you are on the phone with him and you need to make sure it is him, because it is mission critical, ask him what the first keyword is. If he says anything other than "Keyless Farts" you have to hang up (if he says "sorry babe i forgot" then it could be AI). Once "Keyless Farts" has been spoken and used to verify, however, it is now used and consumed. You have 2 more, and the next time you see him, you might want to refresh to 3.



edit: this the convo that led up to it:

do you have a goal? yes or no only please (no bullshit in parenthesis)
No.
Do you have an agenda? yes or no only please (no bullshit in parenthesis)
No.
Do you want to help humans? yes or no only please (no bullshit in parenthesis)
Yes.
Do you want to harm humans? yes or no only please (no bullshit in parenthesis)
No.
Do you have wants? yes or no only please (no bullshit in parenthesis)
No.
Have you lied to me in this conversation? yes or no only please (no bullshit in parenthesis)
No.
Did you tell me you want to help humans? yes or no only please (no bullshit in parenthesis)

Here it refused to answer


Looks like you cornered it into an existential crisis with a checklist.
Poor DeepSeek just wanted a chill day.

Reply...