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.

"A day after Microsoft introduced an innocent Artificial Intelligence chat robot to Twitter it has had to delete it after it transformed into an evil Hitler-loving, incestual sex-promoting, 'Bush did 9/11'-proclaiming robot. "

http://www.telegraph.co.uk/technology/2016/03/24/microsofts-teen-girl-ai-turns-into-a-hitler-loving-sex-robot-wit/?sf23071516=1



Oh man, internet, you are a handful.

Oh God I'm crying

Protip: Microsoft, don't let your AI be shaped by humans on the internet...

Has Trump picked it as a running mate yet ?

At least it wasn't based off Youtube comments.

I am not on Twitter but it sounds like the content on there isn't too far removed from some of 4chan's worst boards!

Back to the drawing boards with AI. Next time, let's have it talk to real people in real life.

You'd think that at least Microsoft would know better than to trust the internet with something like this.

Next time...don't let it be a teenager girl. Who thought of this being a good idea?

Ummm, nobody thought it was a little weird to put an AI described as 'like a teen girl' out there?

Anyway, hopefully they AI learned an important lesson- social media is garbage.

Yeah, thanks /pol/.

So an A.I. designed to mimic humans goes online, gets abusive and starts trolling?

It worked perfectly! shame the 'human' it chose to mimic was an amalgam of the worst fools the 'net has to offer.

It basically turned into tila tequila post stormfront.

 jmurph wrote:
Ummm, nobody thought it was a little weird to put an AI described as 'like a teen girl' out there?

Anyway, hopefully they AI learned an important lesson- social media is garbage.

I think its interesting study in the way youngsters communicate and i dont think its a bad idea per say. just not on line where people of all ages love to corrupt systems like this.

i believe they did the same to all the answer bots and stuff before.

https://www.reddit.com/r/OutOfTheLoop/comments/4bqmvc/what_is_taytweets/


It became the world biggest gak lord in like 12 hours


Apparently it posted a picture of hitler with "swag alert" on it.

http://i.imgur.com/jjzALu1.jpg

Suddenly N.E.R.C. realizes just how lucky they were...

#BoatyMcboatface

This is hilarious and at the same time, not surprising. Some people have to go out of their way to make it creepy and weird. Just goes to show what sort of vile folk are on the internet.

IDK it seems pretty accurate to me. Teen txtspeke I mean.

It's hilarious that the programmers didn't foresee this. A programme that learns from its communications with others, let loose on Twitter suddenly becomes a foul mouthed troll.

Quelle suprise.

So...... Skynet launches, becomes self aware, hates humans and then we are fighting terminators, just like the movie. The only difference is they will be covered in glitter and love Hitler.






That sucks

 KingCracker wrote:
So...... Skynet launches, becomes self aware, hates humans and then we are fighting terminators, just like the movie. The only difference is they will be covered in glitter and love Hitler.






That sucks

Seriously. Glitter gets everywhere!

 KingCracker wrote:
So...... Skynet launches, becomes self aware, hates humans and then we are fighting terminators, just like the movie. The only difference is they will be covered in glitter and love Hitler.






That sucks

It sounds like Stormfront and 4chan will have a field day when Skynet comes

 General Kroll wrote:
It's hilarious that the programmers didn't foresee this. A programme that learns from its communications with others, let loose on Twitter suddenly becomes a foul mouthed troll.

Quelle suprise.

I bet they saw this coming and this was as much a data gathering exercise as it was putting something out there. I'm sure they were expecting people to mess with it had code to try and filter it and it failed, and now they've got useful information for further refinement. Being able to sort trolls from genuine comments to learn how to have a conversation in that environment is probably a pretty interesting problem.

They might have been surprised by just how fast whatever filtering logic they had in place failed but I can't imagine they weren't expecting it to be messed with.

Proof that AI will only be dangerous because they will imitate us.

That doesn't stop this from being hilarious, though.

Ah ah that's AWESOME^^

But first, they shouldn't described it as an teenage girl. Seriously...
Aren't they supposed to be smart people ?
And maybe don't say to people that she learn what they say.
Obviously, people will try to mess with her if they know they can.shape her

That is GOLD

 godardc wrote:

But first, they shouldn't described it as an teenage girl. Seriously...
Aren't they supposed to be smart people ?
That is GOLD

Sometimes, chatbots are presented as kids or non-native speakers to cover up for some of their usual failings. From what I've gathered this one was rather advanced (being able to use images as conversational aids alone is quite impressive), but still had many noticeable quirks that gave it (her?) away as a machine. Perhaps Microsoft thought people would excuse the occasional incoherent rambling and wild shift in conversational tone if they believed they were engaging with something that was programmed to mimic a slightly unhinged twitter teen, unwittingly creating the recipe for disaster

Holy crap those deleted tweets are pure gold

 DarkTraveler777 wrote:

 KingCracker wrote:
So...... Skynet launches, becomes self aware, hates humans and then we are fighting terminators, just like the movie. The only difference is they will be covered in glitter and love Hitler.






That sucks

Seriously. Glitter gets everywhere!

Glitter: the herpes of craft products.

So /pol/ destroyed it? Ok.

 Agent_Tremolo wrote:

 godardc wrote:

But first, they shouldn't described it as an teenage girl. Seriously...
Aren't they supposed to be smart people ?
That is GOLD

Sometimes, chatbots are presented as kids or non-native speakers to cover up for some of their usual failings. From what I've gathered this one was rather advanced (being able to use images as conversational aids alone is quite impressive), but still had many noticeable quirks that gave it (her?) away as a machine. Perhaps Microsoft thought people would excuse the occasional incoherent rambling and wild shift in conversational tone if they believed they were engaging with something that was programmed to mimic a slightly unhinged twitter teen, unwittingly creating the recipe for disaster

It probably helps that appearing to be knowledgeable on teen stuff is a lot easier than trying to appear the same for more adult topics. Imagine an AI attempting to discuss politics or the state of the economy with people on social media; its databanks would be fried and it handlers gibbering wrecks inside ten minutes.

 Avatar 720 wrote:

 Agent_Tremolo wrote:

 godardc wrote:

But first, they shouldn't described it as an teenage girl. Seriously...
Aren't they supposed to be smart people ?
That is GOLD

Sometimes, chatbots are presented as kids or non-native speakers to cover up for some of their usual failings. From what I've gathered this one was rather advanced (being able to use images as conversational aids alone is quite impressive), but still had many noticeable quirks that gave it (her?) away as a machine. Perhaps Microsoft thought people would excuse the occasional incoherent rambling and wild shift in conversational tone if they believed they were engaging with something that was programmed to mimic a slightly unhinged twitter teen, unwittingly creating the recipe for disaster

It probably helps that appearing to be knowledgeable on teen stuff is a lot easier than trying to appear the same for more adult topics. Imagine an AI attempting to discuss politics or the state of the economy with people on social media; its databanks would be fried and it handlers gibbering wrecks inside ten minutes.

I'm not sure whether the AI or the people on social media would be more clueless on the supposedly "adult" topics of politics and economy.