Is A.I. waiting to take over ?

[Foreigner]

Current AI models are not sentient. They are just really elaborate statistics and information collating machines. They can network many different ideas together and summarize data for us, but they can't tell us the value of it. People should watch more sci-fi. The computers in shows like Star Trek are complex assistants to humans. Calculations that would take us days (or years) can be done by a computer instantly, and they will show us whatever we ask them to show us. But they can't tell us if the information is valuable, if it's good or evil, if it can or should be used, or ethics.

For example, there are millions of PhD level articles out there in the journals... but no human being can ever sit down, read them all, and then tell us what unique things they figured out by reading so many articles. It would take many lifetimes. An AI could though. It could read an entire publication and tell us what it found. This would be really useful for things like medicine. An AI could read all of the medical articles ever written and network ideas together in ways we never considered, giving us new treatment options.

That's the difference between spider indexing, like what Google does with websites, an AI. An AI could read all of Bluelight and tell us whatever we want to know about it -- but it wouldn't be able to tell us what is "good" or "bad" about it.

AI could work to our advantage if we could train it to be fair. But we are already seeing major biases being built into these models, like Google's piece of shit Gemini AI. Corporations are already looking at how to game AI.

If you're not a critical thinker, then AI will just lead to way, way worse echo chambers... because AI is only going to ever tell you what you ask if. You don't know how to word your questions fairly, you're going to get biased answers. And in the case of garbage AI like Gemini, the answers will be skewed no matter what you do.

 

[manicmiracles]

Okay, without reading anything and bordering on a Valium rant - yes and no. Trust me bro, I'm from the internet and may have written one or two for my jobs.

The immediate ramp up of speed to where we'd be aware of what is happening would be super small. AIs need to learn to combine encodings, manipulate the physical world, and many other things. Without some eithics part of its weights, of course it would be the superior and dominant species.

 

[snmfmy]

That's insane.

Things are moving way too fast for me to keep up. Maybe the digital age will just become an obsolete ghost town and humans will be divided into decision makers and slaves.

I don't know. We need exit plans. We need a location to retreat to in case everything gets fucked up.

It's the technological singularity. It's a phrase or a term coined by Ray kurtzweiler. It means the point at which technological advancement proceeds to exponentially increase and it is useless to attempt to predict future technological advancement.

Think of it as the event horizon of technological progress. Things are progressing so quickly. It's useless to look into the future.

 

[snmfmy]

Dude if AI makes us live forever I want out

Or I'll go back on meth.

There's a small probability that is increasingly every day that technological advancement will result in functional immortality for people that are still alive now.

 

[snmfmy]

Current AI models are not sentient. They are just really elaborate statistics and information collating machines. They can network many different ideas together and summarize data for us, but they can't tell us the value of it. People should watch more sci-fi. The computers in shows like Star Trek are complex assistants to humans. Calculations that would take us days (or years) can be done by a computer instantly, and they will show us whatever we ask them to show us. But they can't tell us if the information is valuable, if it's good or evil, if it can or should be used, or ethics.

For example, there are millions of PhD level articles out there in the journals... but no human being can ever sit down, read them all, and then tell us what unique things they figured out by reading so many articles. It would take many lifetimes. An AI could though. It could read an entire publication and tell us what it found. This would be really useful for things like medicine. An AI could read all of the medical articles ever written and network ideas together in ways we never considered, giving us new treatment options.

That's the difference between spider indexing, like what Google does with websites, an AI. An AI could read all of Bluelight and tell us whatever we want to know about it -- but it wouldn't be able to tell us what is "good" or "bad" about it.

AI could work to our advantage if we could train it to be fair. But we are already seeing major biases being built into these models, like Google's piece of shit Gemini AI. Corporations are already looking at how to game AI.

If you're not a critical thinker, then AI will just lead to way, way worse echo chambers... because AI is only going to ever tell you what you ask if. You don't know how to word your questions fairly, you're going to get biased answers. And in the case of garbage AI like Gemini, the answers will be skewed no matter what you do.

An AI can absolutely tell us what would be good and bad after reading through every single bluelight post.

Good and bad will be defined by whatever we define it as in the code.

That's the thing with computers and it's the same with people we are told the definition of good and bad from the time we are children. Good and bad are subjective and are different with respect to almost every person on the planet. It is extremely difficult for human beings to edit or rewrite the code that defines good and bad after they've learned it as a child.

And that's the best way to look at computers and AI as children that haven't been taught how to be a human being.

If we told an AI find all of the posts that were bad and we defined bad as containing a curse word. An AI would be able to tell us all about it.

 

[snmfmy]

There's also a whole lot of sensitivity regarding voltage, waveform, interaction with trace metals in the protoplasm surrounding the neurons that apparently gives another order of magnitude to data storage capabilities of the brain.

I mean this is on the very edge of figuring out exactly what goes on in her brain when we perceive something with our sensorium and obviously once completed it will allow us to create the matrix.

 

[Deleted member 521610]

AI is coming, and its going to replace humanity one day.

 

[ĐmantuNJofra]

I'm not tech-wiz but aren't all those ais are LLMs and the only smart ones are self driving combine harvesters?

 

[snmfmy]

@snmfmy Don’t you think AI is a serious matter of National Security? Do you really think the NSA/CIA are sitting around watching this AI rollout thinking “I wonder if it’s going to take over the World someday and kill us all”

Or do you think they’ve developed ASI that is already in control of everything?

Also, you don’t think the CIA would have developed some ChatGPT-like AI a decade ago? I mean think of all the data the CIA/NSA has access to. It wouldn’t be too hard to develop a vastly superior ChatGPT AI with the absolute gold mine of data those agencies have access to.

Independent action outside of things that are define in code doesn't happen.

So no, they're not worried

 

[FindingSobriety]

Independent action outside of things that are define in code doesn't happen.

So no, they're not worried

Correct me if I’m wrong, but even the creators of these LLM’s don’t exactly know how they work, right? Isn’t their “code” sort of a mystery at this point?

 

[snmfmy]

@snmfmy Don’t you think AI is a serious matter of National Security? Do you really think the NSA/CIA are sitting around watching this AI rollout thinking “I wonder if it’s going to take over the World someday and kill us all”

Or do you think they’ve developed ASI that is already in control of everything?

Also, you don’t think the CIA would have developed some ChatGPT-like AI a decade ago? I mean think of all the data the CIA/NSA has access to. It wouldn’t be too hard to develop a vastly superior ChatGPT AI with the absolute gold mine of data those agencies have access to.

Regarding the development of a chat GPT 'like' AI by the CIA, and your assertion that the CIA/NSA has access to some enormous amount of data:

Both are unfounded based on reality.

Regardless of what you may see on TV, the CIA is not home to a whole bunch of Uber capable hackers that control the strings of advanced technology computer programming.

Neither does the NSA have access to what people think is 'all the data flow'.

NSA may have access to a lot of after the fact of metadata.

However, the way that cell networks run today and the absolute absence of data recording and duplication tabs at every single one of the hundreds of thousands of cell towers and repeaters means they don't have real time access to the overwhelming majority of communications in the world, nor do they have the capability to capture the data in a message or a phone conversation in the overwhelming majority of cases.

This is simply because of network topology and the absolute necessity to decrease latency at every single node.

Consistent sub 30 millisecond latencies are impossible to maintain if you're duplicating and recording the packets that are running through a router with any significant amount of traffic.

Much less being able to re-transmit them in real time cuz now you're doubling the actual data throughput which isn't going to happen.

I mean I can sit here and explain things in depth but really there's no point.

There's literally too many people. Too many cell phones. Too many computers. Too many connections to capture, analyze, assess anything more than a small fraction of daily communications.

In the same way that 210,000 vehicles cross our southern border through a valid port of entry every single day, which means at best we could search 10% of them. If we spent the resources of every single employee in customs and border patrol, the NSA and CIA can't listen to, record, or duplicate anything more than a small percentage of communications.

This is simply because the data in streaming applications like only fans, Chaturbate, zoom, Google video calls, and Apple video chat etc and call, telephone calls, etc everyday exceeds the entire data storage capacity on the planet.

Voice telephone calls consist of around a megabyte per minute. Everyday there's about 20 billion telephone calls of varying length.

1 hour of streaming video uses three gigabytes of data.

And realize, that you can't record a video call without having the router. Duplicate the packets that constitute the data. That means that you would be artificially doubling the data flow and that just ain't happening.

 

[snmfmy]

Correct me if I’m wrong, but even the creators of these LLM’s don’t exactly know how they work, right? Isn’t their “code” sort of a mystery at this point?

No, you're wrong. They know exactly how the code works in every single piece of computing code that's out there.

Because regardless of if a program can self-modify code which some can, they can only solve modify code within the constraints of the language and the operating system and the hardware constraints of the platform the code is operating on.

This is a finite environment. There are no question marks.

 

[FindingSobriety]

Regarding the development of a chat GPT 'like' AI by the CIA, and your assertion that the CIA/NSA has access to some enormous amount of data:

Both are unfounded based on reality.

Regardless of what you may see on TV, the CIA is not home to a whole bunch of Uber capable hackers that control the strings of advanced technology computer programming.

Neither does the NSA have access to what people think is 'all the data flow'.

NSA may have access to a lot of after the fact of metadata.

However, the way that cell networks run today and the absolute absence of data recording and duplication tabs at every single one of the hundreds of thousands of cell towers and repeaters means they don't have real time access to the overwhelming majority of communications in the world, nor do they have the capability to capture the data in a message or a phone conversation in the overwhelming majority of cases.

This is simply because of network topology and the absolute necessity to decrease latency at every single node.

Consistent sub 30 millisecond latencies are impossible to maintain if you're duplicating and recording the packets that are running through a router with any significant amount of traffic.

Much less being able to re-transmit them in real time cuz now you're doubling the actual data throughput which isn't going to happen.

I mean I can sit here and explain things in depth but really there's no point.

There's literally too many people. Too many cell phones. Too many computers. Too many connections to capture, analyze, assess anything more than a small fraction of daily communications.

In the same way that 210,000 vehicles cross our southern border through a valid port of entry every single day, which means at best we could search 10% of them. If we spent the resources of every single employee in customs and border patrol, the NSA and CIA can't listen to, record, or duplicate anything more than a small percentage of communications.

This is simply because the data in streaming applications like only fans, Chaturbate, zoom, Google video calls, and Apple video chat etc and call, telephone calls, etc everyday exceeds the entire data storage capacity on the planet.

Voice telephone calls consist of around a megabyte per minute. Everyday there's about 20 billion telephone calls of varying length.

1 hour of streaming video uses three gigabytes of data.

And realize, that you can't record a video call without having the router. Duplicate the packets that constitute the data. That means that you would be artificially doubling the data flow and that just ain't happening.

But the CIA and NSA definitely do have similar capabilities when compared to the Chinese/Russians, right? (Like SNAKE and the like).

Wouldn’t you agree that everything the Chinese/Russian intelligence assets can do we (NSA/CIA) can do? (Or do better).

If the Chinese/Russians can hack into big Corporations, surely we can as well?

 

[FindingSobriety]

No, you're wrong. They know exactly how the code works in every single piece of computing code that's out there.

Because regardless of if a program can self-modify code which some can, they can only solve modify code within the constraints of the language and the operating system and the hardware constraints of the platform the code is operating on.

This is a finite environment. There are no question marks.

Well, you certainly seem to know what the fuck you’re talking about, so I trust your word.

 

[FindingSobriety]

However, the way that cell networks run today and the absolute absence of data recording and duplication tabs at every single one of the hundreds of thousands of cell towers and repeaters means they don't have real time access to the overwhelming majority of communications in the world, nor do they have the capability to capture the data in a message or a phone conversation in the overwhelming majority of cases.

Key words being “Majority of cases”

You know just as well as I do we have access to programs like Pegasus and the like which allow us to do all that shit (and more) with the press of a button. You’re basically saying that type of surveillance isn’t implementable on a broader scale, right?

 

[PtahTek]

I would say "they" scrape everything. The tech is undeniable so why not?

 

[Jabberwocky]

Yeah I really wouldn't be surprised by anything at this point. Personally I'm just trying to plan on getting in front of the movement so that I'm not swept away by it.

At the moment it seems harmless enough.

However I know that its only going to get more dangerous.

 

[snmfmy]

Key words being “Majority of cases”

You know just as well as I do we have access to programs like Pegasus and the like which allow us to do all that shit (and more) with the press of a button. You’re basically saying that type of surveillance isn’t implementable on a broader scale, right?

That would be correct.

Pegasus can't be officially used against any US citizens on US soil by any US law enforcement agency, local, state, or federal. President Biden blacklisted the use of NSO's technology.

Any information gained through the use of Pegasus would be fruit of a poisoned tree and could not be used.

And yes, it is a limited deployment because of the prohibitive licensing fee and there's a bottleneck with respect to NSO servers.

 

[Jabberwocky]

Hey guys just realized that this would be better suited for current events forum. Interesting conversation - I'll leave a redirect thread in philosophy subforum so both are visible.

 

[snmfmy]

AI is coming, and its going to replace humanity one day.

Humanity will become transcendent humanity where those of us who decide to become post-human or transhuman will protect all of our cousins who decide to be luddites and not jump on the post-humanity bandwagon.

I really want to be alive when functional immortality or the ability to partially upload etc. Gives us lifespans on the order of eons because I want to see what's in the universe and right now it really doesn't look like we're going to be able to have Star Trek warp drive and at a reasonable extension of technology. It would take 400,000 years to cross our galaxy (That's at 25% speed of light and the galaxy is 100,000 light years in diameter, Yes, there would be some time dilation but not as much as people think. It might seem a little shorter than 400,000 years, but to everybody else in the galaxy it would be 400,000 years) never mind stopping off on the way.

And the scenic tour to Andromeda would be 10 million years just to get there considering it's 2.3 million light years away. And you multiply by four since you're going 25% SOL (isn't that funny that's the name of the sun. That means shit out of luck and also is a acronym for speed of light?).

Now I would like to believe that we will be able to artificially create enough virtual mass to collapse space time and when I say collapse, I mean compress warp make it shrink in front of us and expand it behind us so that our little bubble of spacetime isn't moving any faster than the speed of light. In fact it we're not moving at all, but space time is moving through itself like a slinky and there's no limit on that. Modified spatial distortion inchworm drive first thought up by Miguel Alcubierre. And his math checks out it should work. We just don't know how to go about creating a multi-jupiter mass and an area a few hundred kilometers across directly in our direction of travel since we'll be falling towards it and then magically make it go away and push from behind us.

The math says we should be able to do it. And if the technological singularity hits soon enough, that means we'll have the time to do it.