Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

I remember reading about how AI will never become self aware in wh40k but I dont remember where. It was actually a really easy explanation. Does anyone recall? What is the general view regarding Ai taking over?

AI already became self-aware in the form of the Men of Iron. I read an interesting theory about how the Machine Spirit is actually the remnants of a code beamed into space by the Men of Iron at the end of their rebellion.

AI will never become self aware because all computers/mechanical machines require programming (orders) to be input manually by a techpriest. They only do what the program tells them to do and nothing else.

 RLRAFTERMAN wrote:
AI will never become self aware because all computers/mechanical machines require programming (orders) to be input manually by a techpriest. They only do what the program tells them to do and nothing else.

isnt that all machines? At some point though machines learn to automate processes. My question is why in other sci fi realms do machines become self aware and turn against humans, yet in wh40k they seem like simple machines that need an operator.

I cant explain as to the other sci-fi realms but for 40k they make them require an operator constantly feeding it orders so they do not become self aware.

 RLRAFTERMAN wrote:
I cant explain as to the other sci-fi realms but for 40k they make them require an operator constantly feeding it orders so they do not become self aware.

have they atleast addressed this issue in the fluff?

This is from the wiki

Artificial intelligence (also refered to as "Abominable Intelligence" or "Silica Animus" in the Imperium) is prohibited under pain of execution by Imperial edict within the Imperium of Man unless a human mind is combined with it or serves as part of its mechanism (hence the existence of Servitors). This point was made part of the agreement between the Imperium and the Adeptus Mechanicus from the signing of the Treaty of Mars in the 30th Millennium, to ensure that intelligent machines would never again become corrupted by Chaos and rebel against their human masters, as happened during the Dark Age of Technology with the constructs known as the Men of Iron, who launched a star-spanning rebellion against their human creators.

 chuckz1487 wrote:

 RLRAFTERMAN wrote:
AI will never become self aware because all computers/mechanical machines require programming (orders) to be input manually by a techpriest. They only do what the program tells them to do and nothing else.

isnt that all machines? At some point though machines learn to automate processes. My question is why in other sci fi realms do machines become self aware and turn against humans, yet in wh40k they seem like simple machines that need an operator.

No it a old school concept from before the age of built in memory. There is no way for it to automate a processes as it can not remember what it did let alone how.

Noir wrote:

 chuckz1487 wrote:

 RLRAFTERMAN wrote:
AI will never become self aware because all computers/mechanical machines require programming (orders) to be input manually by a techpriest. They only do what the program tells them to do and nothing else.

isnt that all machines? At some point though machines learn to automate processes. My question is why in other sci fi realms do machines become self aware and turn against humans, yet in wh40k they seem like simple machines that need an operator.

No it a old school concept from before the age of built in memory. There is no way for it to automate a processes as it can not remember what it did let alone how.

I disagree... a computer knows how to start up, load your profiles, load your start up programs, and respond to certain situations (viruses, blocking pop ups). I would assume that machines in the 40k would be more intelligent than that. Just like a computer knows how to clean up trash junk files, why couldnt a machine "get rid" of a human who it views it as inefficient?

I want to say that it was in a Horus Heresy novel that explained it.... damn i cant remember

http://wh40k.lexicanum.com/wiki/Silica_Animus

Read: The Death of Integrity

That is why self aware AI are a bad thing in WH40K

 chuckz1487 wrote:

Noir wrote:

 chuckz1487 wrote:

 RLRAFTERMAN wrote:
AI will never become self aware because all computers/mechanical machines require programming (orders) to be input manually by a techpriest. They only do what the program tells them to do and nothing else.

isnt that all machines? At some point though machines learn to automate processes. My question is why in other sci fi realms do machines become self aware and turn against humans, yet in wh40k they seem like simple machines that need an operator.

No it a old school concept from before the age of built in memory. There is no way for it to automate a processes as it can not remember what it did let alone how.

I disagree... a computer knows how to start up, load your profiles, load your start up programs, and respond to certain situations (viruses, blocking pop ups). I would assume that machines in the 40k would be more intelligent than that. Just like a computer knows how to clean up trash junk files, why couldnt a machine "get rid" of a human who it views it as inefficient?

I want to say that it was in a Horus Heresy novel that explained it.... damn i cant remember

That just it the IoM doesn't use that type of computer. We are looking at Tandy computer a super advanced Tandy but a Tandy nonetheless. There is no profile to start it working, you need to give it the code for the program you want to run, and maybe a dataslate if the program is not on the machine. Sure they could be more advanced but they have already seen the outcome and now would never give a machine that much freedom.

 chuckz1487 wrote:

 RLRAFTERMAN wrote:
AI will never become self aware because all computers/mechanical machines require programming (orders) to be input manually by a techpriest. They only do what the program tells them to do and nothing else.

isnt that all machines? At some point though machines learn to automate processes. My question is why in other sci fi realms do machines become self aware and turn against humans, yet in wh40k they seem like simple machines that need an operator.

Because that already happened during the Dark Age of Technology.

Humans created a race of sentient robots called the Men of Iron to serve them(they were basically Terminators). They rebelled and nearly drove humanity to extinction, mankind won. Just barely.

Every since, the human psyche has had a massive and healthy fear of technology. The Adeptus Mechanicus's doctrine holds that a true self-aware AI is an abomination, they remember what happened in the Dark Ages.

So in place of AI, they have the machine spirit. A cross between advanced computational algorithims(little different from our computers today in basic function) and human nervous tissue. By making their AI substitute rely on humans both for input and construction, and regularly wiping system data, they ensure that a sentient AI never appears again.

So the Imperium has what we today would call AI, but they are not sentient. They're just really advanced computers.

 chuckz1487 wrote:

Noir wrote:

 chuckz1487 wrote:

 RLRAFTERMAN wrote:
AI will never become self aware because all computers/mechanical machines require programming (orders) to be input manually by a techpriest. They only do what the program tells them to do and nothing else.

isnt that all machines? At some point though machines learn to automate processes. My question is why in other sci fi realms do machines become self aware and turn against humans, yet in wh40k they seem like simple machines that need an operator.

No it a old school concept from before the age of built in memory. There is no way for it to automate a processes as it can not remember what it did let alone how.

I disagree... a computer knows how to start up, load your profiles, load your start up programs, and respond to certain situations (viruses, blocking pop ups). I would assume that machines in the 40k would be more intelligent than that. Just like a computer knows how to clean up trash junk files, why couldnt a machine "get rid" of a human who it views it as inefficient?

I want to say that it was in a Horus Heresy novel that explained it.... damn i cant remember

A PC does not know to clean up junk files without someone telling it to. Your Recycle Bin is not self-cleaning. Your browser-cookies are not self-deleting. Your TMP folders are not self-deleting.

There are applications you can install/configure to tell your computer to do this, but this, again, has required operator input in order to enable this function. Remove those apps, and their functions stop happening. Your firewall comes pre-configured to stop certain types of virus and malware attacks... it requires a human programmer to update its code in order to address new threats, human programmers must code OS updates to close security flaws, hackers must code new viruses to attack newly-found exploits. These are all things that require a human agent to take action, the computers do not do them on their own.

I like to think that the 'computers' of 40k are more analogous to those of towards the end of the space race(flashing lights, dials, etc) than anything we think of when we say computer today.

Thinking about it, for me 40k is basically 'The Past, but in the future and with space lasers!

They had AI.
It rebelled.

Almost wiped the pinkies out.

If there's one thing the imperium is, it's incredibly superstitious. They don't like to tempt "fate" - so they don't try to get anywhere near that level again. The burned hand teaches the best and all that.

So after fighting back from the brink, they outlawed it. Not unlike what they did in the Dune universe (which is why they trained human minds to be their cogitators - the mentats ). Same reason.

Stands to reason, really. Dune is one of the sources they homaged when they made the game.

nareik wrote:
I like to think that the 'computers' of 40k are more analogous to those of towards the end of the space race(flashing lights, dials, etc) than anything we think of when we say computer today.

Thinking about it, for me 40k is basically 'The Past, but in the future and with space lasers!

No, their computers are far beyond anything we have today. They are just illustrated as being large and clunky because that's the aesthetic of the setting.

The Tau are progressing pretty quickly toward self-aware A.I, IMO. I could actually see them pulling it off without the usual repercussions, too, given their fluff exploits.

On top of that - I imagine there's very little that is technologically beyond the Necrons, and the Eldar were masters of automation back in their hayday, so it wouldn't surprise me if both of those races could create a self-aware AI too.

 Robin5t wrote:
The Tau are progressing pretty quickly toward self-aware A.I, IMO. I could actually see them pulling it off without the usual repercussions, too, given their fluff exploits.

On top of that - I imagine there's very little that is technologically beyond the Necrons, and the Eldar were masters of automation back in their hayday, so it wouldn't surprise me if both of those races could create a self-aware AI too.

The first thing Tau AI will do is look at the Tau caste structure and see how oppressive the ethereal caste actually is once you wipe away the smoke screen they have created. Rebellion is inevitable.

The first thing Tau AI will do is look at the Tau caste structure and see how oppressive the ethereal caste actually is once you wipe away the smoke screen they have created. Rebellion is inevitable.

That, or see how vulnerable their system really is and simply take it over. It's possible that the Ethereal's could be enslaved by a good enough AI with no one being the wiser. Until it's too late ...

As far as the OP - I don't think I've ever heard/read anything about it being impossible for AI to happen in the 40k Universe. Just that, as others have said, it's outlawed in the IoM. If you read some of the more recent novels, you will see examples where Tau actually already have some drones with limited AI functionality. Not "true" AI mind you, but certainly a level of independent decision making capability/situational awareness that you likely will never see in a mechanicus/IoM machine.

The tau have networked intelligence.
You need a minimum number of members linked in order to qualify.

I don't think they LET enough of them link to get to the "skynet" level.

The tau have networked intelligence.
You need a minimum number of members linked in order to qualify.

I don't think they LET enough of them link to get to the "skynet" level.

I think that's true for most of the drones but there are examples of drones that are slightly smarter than others. For example, in the Damocles novel, Shadowsun's shield drone is not only self aware enough to make independent decisions, it's also smart enough to do things like thank her for dodging a lascannon shot when the drone's energy is too low to absorb it.

 Robin5t wrote:
The Tau are progressing pretty quickly toward self-aware A.I, IMO. I could actually see them pulling it off without the usual repercussions, too, given their fluff exploits.

If they do pull it off surely this will push them higher into the IoM priorities of enemies as Mars will not stand for this tech heresy. They will stop being "Eh, when we get around to them" and become "Number 1, super mega enemy"

I think the collective wraithbone minds that inhabit and control Eldar craftworlds could very well qualify as AIs. They might be based on "uploaded" minds (consciousness? patterns? souls?) originating from biological brains, but they have selfdelveloped very far beyond that.

Maybe someone who is more up to date on Necron fluff have an opinion on if they would qualify as well.

Mellon wrote:
I think the collective wraithbone minds that inhabit and control Eldar craftworlds could very well qualify as AIs. They might be based on "uploaded" minds (consciousness? patterns? souls?) originating from biological brains, but they have selfdelveloped very far beyond that.

Maybe someone who is more up to date on Necron fluff have an opinion on if they would qualify as well.

Isn't Wraithbone not even remotely similar to computers, technically speaking?

It's not AI in any sense, but instead spirits souls and their like?

The infinity matrix is pretty much every single dead Eldar soul from that Craftworld but they lose most of their self awareness except the Seers I think.

 chuckz1487 wrote:
I want to say that it was in a Horus Heresy novel that explained it.... damn i cant remember

The "Kaban Project" ?

http://wh40k.lexicanum.com/wiki/Kaban_Machine

There is an A.I. in "Cybernetica" too. Bad idea to let that abomination run loose to beat the traitors.

And the Castigator-class Titan......

The Soulless sentience is the enemy of all.
The alien mechanism is a perversion of the True Path.

 chuckz1487 wrote:

Noir wrote:

 chuckz1487 wrote:

 RLRAFTERMAN wrote:
AI will never become self aware because all computers/mechanical machines require programming (orders) to be input manually by a techpriest. They only do what the program tells them to do and nothing else.

isnt that all machines? At some point though machines learn to automate processes. My question is why in other sci fi realms do machines become self aware and turn against humans, yet in wh40k they seem like simple machines that need an operator.

No it a old school concept from before the age of built in memory. There is no way for it to automate a processes as it can not remember what it did let alone how.

I disagree... a computer knows how to start up, load your profiles, load your start up programs, and respond to certain situations (viruses, blocking pop ups). I would assume that machines in the 40k would be more intelligent than that. Just like a computer knows how to clean up trash junk files, why couldnt a machine "get rid" of a human who it views it as inefficient?

I want to say that it was in a Horus Heresy novel that explained it.... damn i cant remember

These are run by "scripts", basically they are a to do list that can be setup or modified by the user. The AIs in 40k were probably like the ones in "I, Robot" the butlers, military bots and bots that will make parttime jobs extinct.

Also, technically AI in the Imperium can never become self-aware, becaue it plain does not exist. In a purely mechanical form, i's even more rudimentary than what we have now. Everything that has any "computing" power that can take any self-initiative is due to the biological brain being involved as a servitor. And even those are largely mindless drones.