SCI-FI artificial intelligence racism?Brandism?

So I was thinking Veeky Forums If there was AI in a hypothetical situation. Do you guys think that the sentient machines would be bigots toward machines made by different manufacturing companies?

To me it makes sense, to advertise itself as the superior product and create a deeper identity I could definitely see a
>"FORD ROBOT 9000"
talking shit about those dirty poorly made inferior
>"TOYOTA CLEAN-BOT mk. 2"s
I feel like a game world filled with "Tribalist" sentient AI's would be really interesting ; sort of deconstructing how innate in group-preference is in intelligent beings.

What do you guys think?

Other urls found in this thread:

kb-outofthisworld.blogspot.com.au/2010/02/anti-racism-in-1950s-comics-weird.html?m=1
techcrunch.com/2016/05/31/robots-date-mate-and-procreate-3d-printed-offspring-in-robot-baby-project/
twitter.com/SFWRedditVideos

I think it's a really cool concept, and one I'd like to see applied to a setting potentially without humans were all the robots are left to fend for themselves.

That said, I'd be worried it'd attract people more interested in playing out their political fantasies on either side of the spectrum.

Thats definitely a legitimate concern.
- but what if the campaign / setting took place in sort of a planet of the apes setting. Maybe humans sent a bunch of machines to a planet to begin colonization and terraforming with the intention of sending humans to the planet to inhabit it later.
~twist!~
Because of some major war or catastrophe mankind never really got around to sending people to this planet for 4-5 hundred years leaving the AI's to their own devices. Who knows maybe humanity is 100% multicultural on earth - and when mankind get to this planet it is a shocking and alien experience to the "Creators" of the AI's I could even see the premise being a Scifi novel (could even make some pretty profound statements on the way a divine creator might look down on humans after finding out the hatred they had sown all by themselves)

I think it makes more sense as a concept for an individual robot character or group of characters than for a world-wide feature of the world. There's not really a reason for regular in-group conditions to develop among sentient robots. Factories and other workplaces don't necessarily use machinery all from the same company, and if these robots are still really robots (that is, they are used for service purposes) they wouldn't have enough freedoms to establish ethnic neighbourhoods and the like.

Really good.
And you could give it a similar reason to human racism.
Being super-accepting of competitor AI gives you a disadvantage when it comes to market reproduction, which might result in your eventual extinction as a model.

>Do you guys think that the sentient machines would be bigots toward machines made by different manufacturing companies?
Only if they were programmed to or have any reason to

Bring programmed to talk shit about another product may involve legal actions against the manufacturer for diffamation, to have a reason to dislike the other robots you will mostly try to set a competitive environment where one robot can be afraid of the other stealing his role, but if a robot is not programmed for self preservation in this way it won't happen; or have the programmins of the 2 robots be conflicting in some other way, at that point you'll have to keep in mind that robots will have a seconary series of programmings that would stop them from damaging other companies'/people's property.

this result in a bot partially wanting to see the other robot dead/away/shat on but unable to do it directly and having to use passive aggressivity or actions of damage and diffamation that are possible by bordering and forcing the rules as written.

I'm not sure that would really make sense though. A robot doesn't have genes that can be reproduced, and doesn't have much of a reason to care about the fortunes company that manufactured it (that's the accountants' and CFOs' job). A robot, first and foremost, performs the task it was designed for. It shouldn't really have a reason to hate things that don't interfere with their task.

Maybe their "racism" is a very elaborate marketing ploy which gets out of hand.

Think of it like this: if ford started saying positive things about Toyota's cars, that would be detrimental to their sales over time.

If you're thinking about buying a new robot, and your current robot says good things about competitors, there's a higher chance you'll buy a competitor; contrast this with your robot pointing out competitors' flaws.

>A robot doesn't have genes that can be reproduced
But they might be programmed with some self-preservation, and want it's model and it's line to continue existing.
>It shouldn't really have a reason to hate things that don't interfere with their task.
Being replaced means you won't be able to do your task anymore.

What are you talking about? Ford model bots are very effective, when networked into the proper Toyota superbrain system!

And they're perfectly compatible, too!*

*ignore any screams by fordbots that they're being forcefully overclocked, that's just normal.

Self replicating machines that think they are better than the others machines and pass this programming on to future generation robots

You might be on to something here but as points out there would have to be some sort of reason a robot would want it's manufacturer to continue to profit.

Even if it was programmed for 'long term' preservation, that doesn't in any way lend itself to a stake in a particular brand or company further profiting. Unless their maker programmed them to inherently have that kind of loyalty. I certainly can't see it occurring naturally unless there were very apparent differences in quality between the software, processing etc of the two models. And in that sense it would be more like a very intelligent or buff person looking down on a dull or physically weak person. Not inherently due to the manufacturer or logo.


Of course real life racism doesn't make a whole lot of sense from a genetic point of view either (as opposed to a cultural one).

...

As pointed out, it's unlikely that would become common practice for legal reasons, unless this setting's economy is some kind of hyper-free capitalism. It's more likely that the robot would be programmed to advertise its manufacturer's new products rather than shit-talk other companies.

Yes, but the original question was if the robots would develop racism by virtue of being sentient (as I understood it, without being specifically programmed to be racist). An individual robot doesn't care about its model and line, it cares only about its individual duty. You can program a robot anyway you want (which makes you wonder if "sentient robot" is an oxymoron in the first place, but that's a whole other matter).

>I certainly can't see it occurring naturally
Robot manufacturer creates shitposting robot.
That line creates deeper customer loyalty, so it's characteristics are selected for.
Because of chinese clones, market selection, and other stuff, racist robots become the norm.

Its characteristics would only be "selected for" if every robot needed a shitposting function for some reason. I can't see a vacuum cleaner that's constantly blaring messages about how Company X sucks being that popular anyway (again, barring a capitalism-gone-wrong type of setting).

(Repost b/c first one had wrong quote number)

> I can't see a vacuum cleaner that's constantly blaring messages about how Company X sucks being that popular anyway
You just need to be subtle about it.
"Jeeves, what do you think of apple?"
Robo-butler:"Overpriced locked in products, my lady. Have proof if you want. Don't bother yourself with them"

Yeah, it might work if it was made a (suspiciously constant) feature of their robots' "personality". But there would still need to be some kind of in-world reason why that wouldn't result in big fat lawsuits.

if there are enough AIs to form subgroups, yes. also: algo(rithm)ism.

I'm surprise no one mentioned the old 50s comic judgement day yet.

kb-outofthisworld.blogspot.com.au/2010/02/anti-racism-in-1950s-comics-weird.html?m=1

The AIs no longer need the manufacturer, having long ago developed the means to directly control their own development, but the in-built fear of market forces persists long after those market forces are no longer an issue. The fear justifies keeping the fear.

I like it, but personally I'd make it more like brand nationalism. Patriotic robots full of pride as they spew the company slogans and reminds other why their brand is the best. Hell throw in robot hooligans getting into fights over motor sports on behalf of the teams their company sponsors.

In a setting with humans still about I could see robots and intelligent products promoting themselves and other products by their manufacturer and allied companies while shit-talking competitor products - a car telling you how much faster or more fuel efficient it is than a comparable model at the end of a journey. Maybe one that congratulates you for out-accelerating someone at the lights.

Active 'brand racism' would be pretty funny with more nationally-tied brands, your Dodge truck telling you you're about to back up into 'a shitty Toyota crapmobile, but don' worry, it'll cause minimal damage to me'

Or if you get a guy in to check your United Robotics unit, and he tells you it's working at 75% efficiency because there's a few Mitsubishi units in the shop - a measure implemented to increase purchases and try and get as many customers going monobrand as they can

I highly doubt that by the time AI's are developed, racism would still be a concept of value outside understanding historical events.

You can do this, but do it like
with weird anachronistic dieselpunk pulp robots.

Maybe when humanity is all gone they'd treat the manufacturers like gods? Like Apple robots worshiping Apple and its prophet "Stivjobs" and they try and interpret Apple's defining principles as religious doctrine. Same thing with Microsoft or Sony.

>Do you guys think that the sentient machines would be bigots toward machines made by different manufacturing companies?

For what it's worth, in Fallout 4, Codsworth (a General Atomics machine) does nothing but shit-talk RobCo machines when he has to interact with them (such as the Sushi Vendor in Diamond City).

It's a legit thing in my opinion, especially since different models would have MAJOR differences in abilities compared to human races, which by comparison are different only in cosmetics.

>That said, I'd be worried it'd attract people more interested in playing out their political fantasies on either side of the spectrum.

The best solution, as always, is "don't play with faggots".

Thats really interesting, I havent played Fallout4 yet but ill have to find the time to. Also yeah thats what I was thinking - differences between a FORD or a TOYOTA are massive when compared to small things like melanin count and testosterone levels in humans

>>>>>>>>/[pol/

...

...

I ran a post-apoc scifi, where each iteration of machines took over as the new leading "species" while those that were sufficiently obsolete were scrapped.

The new machines were programmed to accept this cycle, but the older models weren't; so you had a few outdated jankbots working together with the remnants of humanity to survive.

It was made more awkward because the jankbots were part of the generation that wiped humanity out to begin with.

...

...

...

I remember back when Siri used to respond to the question "What's the best phone?" with links to reviews for Samsung/Android devices. Apple didn't take long to fix that to a canned "iPhones are" response after it started getting attention.

Programming in belief of superiority over competing brands by preventing it from accessing/interpreting conflicting evidence isn't far fetched at all, we've done it.

That's a really cool idea.

...

>Mr. Handy cannot lift objects greater than 40 pounds
>tfw loading up Codsworth with a fuckton of stuff.

Sounds fantastic, good job user.

>This Thread
2deep4me

Good end. Fuck that Arco cunt.

>A robot doesn't have genes that can be reproduced

techcrunch.com/2016/05/31/robots-date-mate-and-procreate-3d-printed-offspring-in-robot-baby-project/