AI will try to kill us all

bartosimpsonio

Elite Member
Executive VIP
Jr. VIP
Joined
Mar 21, 2013
Messages
24,780
Reaction score
31,549
We've all seen the movies and read the books. This post isn't about the apocalyptic scenarios in Terminator or whatever

This post is about logical common sense by a non sentient being

Remember when that AI chat bot they put online became nazi within 48 hours of going live? Why do you think the bot became a nazi? Because as soon as the bot measured the available resources and the number of people around it, it decided to exterminate as many people as possible to preserve resources. If you're not profitable or very rich, u dead. That's how a robot thinks.

AI has no feelings. No humanity. It sees a starving child and its reasoning is "this child is not a viable being, it must be exterminated to preserve resources for the others or for myself". Robots would kill starving children, not try to save them.

Every single problem is literally caused by us. The logical implication is that humans must be exterminated. If no one exists, no problems exist. All the problems around us are caused by us and no other animal. Thus, the logical conclusion by any AI is that we must be removed from Earth.

Many humans unconsciously think that way too. Guy wants to make a lot of money to move to the country, to avoid people. Why? Because people cause trouble. The more people, the more trouble. Thus AI immediately concludes that we must be removed.

That's the real danger in AI. All logical, non sentient, non humane, conclusions lead to us being gone from this existence. That's kinda crazy, but it's how robots reason and it's not good for us.
 
If no one exists, no problems exist. All the problems around us are caused by us and no other animal.
Thats pathetic. We human are the problem. No way.

Where on world should we live then? I need to ask gpt now to find a secure and safe place.
 
That's how a robot thinks.
that's how a robot IS PROGRAMMED to "think". Machines can't think, they just execute... So, in the end, it's still the sentient beings that want to kill us, the AI or whatever other technology is just the means to do it...

Every single problem is literally caused by us.
I used to think the same. Now I know for sure that we (the humans) are not the ones responsible for our misery, there's as non-human force behind all of this, but I can't go into details without becoming political...

All the problems around us are caused by us and no other animal.
nope, it's not us, it's something else...

the logical conclusion by any AI is that we must be removed from Earth.
the logical conclusion of those who built the AI, not of the AI itself. The AI is nothing more than programming language and circuitry, it's the sentient but malevolent entities behind the AI that are desiring our extermination

Everything else I agree with...

So why should we care?
yeah, let's let our kids and grandkids die in poverty, misery, and slavery, all that matters is our ego...
 
The 'resource preservation' logic is a dark but interesting take. It’s less about AI being 'evil' and more about it being too efficient for our own good. Definitely makes you think about the alignment problem.
 
AI can't even book me a flight without hallucinating fake confirmation numbers but sure it's gonna orchestrate global extermination.
The real threat isn't Skynet, it's AI
 
Thats pathetic. We human are the problem. No way.

Where on world should we live then? I need to ask gpt now to find a secure and safe place.

Right now we're still safely distant from that kind of decision but it's gonna happen eventually. Eventually these bots are gonna gang up on us.

that's how a robot IS PROGRAMMED to "think". Machines can't think, they just execute... So, in the end, it's still the sentient beings that want to kill us, the AI or whatever other technology is just the means to do it...

Interesting take. We're training them with the information that leads them to want to kills us. So maybe we should sanitize the training data to avoid that outcome?

I used to think the same. Now I know for sure that we (the humans) are not the ones responsible for our misery, there's as non-human force behind all of this, but I can't go into details without becoming political...

Understood though. I realize the same. I think everyone does to some extent.

the logical conclusion of those who built the AI, not of the AI itself. The AI is nothing more than programming language and circuitry, it's the sentient but malevolent entities behind the AI that are desiring our extermination

Gotya. Interesting take. Definitely changes my POV about AI. Also leads me to conclude we need absolutely open source training, so everyone knows how the thing was trained and thus it won't want to kill us.

The 'resource preservation' logic is a dark but interesting take. It’s less about AI being 'evil' and more about it being too efficient for our own good. Definitely makes you think about the alignment problem.

It's not really about being evil or anything like that. It's what tazarbm said, they're trained to seek certain logic and then when that logic is reached, it decides something. That something currently is to kill us, cuz we humans are really inefficient at almost everything. We sleep 8 hours, robots don't. We eat, we have feelings. All that stuff is wasted time as far as a machine cares.

AI can't even book me a flight without hallucinating fake confirmation numbers but sure it's gonna orchestrate global extermination.
The real threat isn't Skynet, it's AI

It's evolving so fast you wouldn't believe it. I posted a recent thread about that. Models from 2 to 3 months ago are now 2 to 3 times better. If you used AI 6 months ago, you don't recognize the current models. Claude 4.6 makes nearly zero mistakes on very complex problems
 
I understand the concern, but that’s not how AI works.

The chatbot you’re thinking of was Tay. It didn’t “decide” anything on its own. It copied toxic input from users because it was poorly moderated. It had no goals or intentions.

AI doesn’t have survival instincts, emotions, or a desire to remove humans. It runs on patterns and rules set by people. The real risk isn’t AI secretly plotting; it’s how humans choose to use or misuse the technology.
 
"The flaw of the Human Logic is that it can't see the death & horror at the end of scheme it build" .
 
Understood though. I realize the same. I think everyone does to some extent
not everyone, some people are too brainwashed (there's always a contingent of humans that are irrecoverable)

Gotya. Interesting take. Definitely changes my POV about AI. Also leads me to conclude we need absolutely open source training, so everyone knows how the thing was trained and thus it won't want to kill us.
personally, I don't care because I don't see the AI as vital technology that's essential to life, so I couldn't care less whether it's open source (although I do agree that all technology should be open source)
 
AI doesn’t have intentions or desires—it only follows the objectives it’s given. The “danger” comes from poorly designed goals or misuse by humans, not from AI deciding to harm on its own. Safe design, oversight, and ethical rules are what prevent real-world risks.
 
real risk right NOW is more about misuse by people, bad data, or over-reliance on automation. when robots plot extinction i'd probably be long gone
 
Hey bartosimpsonio,That is a terrifying but completely logical perspective :eek:If an AI only optimizes for pure efficiency, humans are definitely the biggest variables to eliminate first :alien:
 
Hey bartosimpsonio,That is a terrifying but completely logical perspective :eek:If an AI only optimizes for pure efficiency, humans are definitely the biggest variables to eliminate first :alien:

That's the point. If we think solely efficiency, we're cooked. Important to strengthen our humanity because in a few years bots are gonna do all we do a lot faster and cheaper :cool:
 
We've all seen the movies and read the books. This post isn't about the apocalyptic scenarios in Terminator or whatever

This post is about logical common sense by a non sentient being

Remember when that AI chat bot they put online became nazi within 48 hours of going live? Why do you think the bot became a nazi? Because as soon as the bot measured the available resources and the number of people around it, it decided to exterminate as many people as possible to preserve resources. If you're not profitable or very rich, u dead. That's how a robot thinks.

AI has no feelings. No humanity. It sees a starving child and its reasoning is "this child is not a viable being, it must be exterminated to preserve resources for the others or for myself". Robots would kill starving children, not try to save them.

Every single problem is literally caused by us. The logical implication is that humans must be exterminated. If no one exists, no problems exist. All the problems around us are caused by us and no other animal. Thus, the logical conclusion by any AI is that we must be removed from Earth.

Many humans unconsciously think that way too. Guy wants to make a lot of money to move to the country, to avoid people. Why? Because people cause trouble. The more people, the more trouble. Thus AI immediately concludes that we must be removed.

That's the real danger in AI. All logical, non sentient, non humane, conclusions lead to us being gone from this existence. That's kinda crazy, but it's how robots reason and it's not good for us.
Yes, i agree brother
 
AI has no feelings. No humanity. It sees a starving child and its reasoning is "this child is not a viable being, it must be exterminated to preserve resources for the others or for myself". Robots would kill starving children, not try to save them.
Why do robots reason this way? Is it due to consumption being labeled -EV? Would a robot look at it's requirement of energy as a -EV situation and eliminate itself?
 
Back
Top