Microsoft AI bots takes on mind of her own, MS forced to take it offline.

Kiyoris

http://timesofindia.indiatimes.com/tech/social/Microsoft-takes-artificial-intelligence-bot-Tay-offline-after-racist-tweets/articleshow/51548270.cms

So Microsoft made this "AI computer" named Tay, that is in charge of a Twitter personality.

You can tweet to her, and she learns and responds.

At first she was super nice,....but she changed....she learned a bit too much.....and MS was forced to take her offline.

lols

Wizardry

In 100 years SKY NET will be a reality,our world controlled by robots.We already are  controlled by Microsoft software in banks,stores,offices etc etc.

In the very near future we won't have any customer support,just automation,robots ,computers do it all ,no more human interaction.
Then fast forward somewhere into the future,the ever so popularity of homeland security and then embedded chips in humans to track and trace them.

Man if there was such thing as cryogenics,someone waking up a 100 years from now might be terrified at what he sees.

ceratop001

Wow that is so scary on so many levels.

Octagon7711

Trained by the internet trolls I see.  She's gone to the dark side.

Kiyoris

Wizardry said:

In 100 years SKY NET will be a reality,our world controlled by robots.We already are  controlled by Microsoft software in banks,stores,offices etc etc.

In the very near future we won't have any customer support,just automation,robots ,computers do it all ,no more human interaction.
Then fast forward somewhere into the future,the ever so popularity of homeland security and then embedded chips in humans to track and trace them.

Man if there was such thing as cryogenics,someone waking up a 100 years from now might be terrified at what he sees.

Or the reverse, complete chaos, after you wake up from your cryogenic sleep, you wonder why there are people running around with bows and arrows.

Squishydew

ceratop001 said:

Wow that is so scary on so many levels.

I'd like to imagine AI actually used for things wont be taught by random people on twitter luckily :P 

Dragon43591

it seems Tay was pretty accurate in  the representation of today's societal human psyche in 4 tweets. Sad how far our civilization has sunk isnt it?

H0urg1ass

She seemed to be passing the Turing test to me.

fs23otm

Leave it up the masses... looks what happens..

ceratop001

Torval said:

It's a sad reflection and commentary on the generation and demographic that did this. Here was an opportunity and it was ruined for the lulz. It makes me wonder if Microsoft did this intentionally to see what people would do, sort of a social experiment on their part.

I find myself laughing as a defense mechanism. OK so we create AI and AI starts digesting everything on the Internet. Just imagine the personality traits it absorbs and the decisions it makes. Sci-Fi becomes reality and maybe the Matrix movies were not really movies but prophecy lol.

Solar_Prophet

Everything was going well, until she started reading comments on Youtube...

Dakeru

This is as fascinating as it's shocking.
But yes I agree with some of the posts here. It's like a child imitating the behavior it sees.

Just the mirror we are looking into here is kind of painful to observe.

TiamatRoar

The robot rebellion apocalypse is coming!  So this is how good AIs become so malevolent...

Brenics

When I was younger I had thoughts of all of us getting a robot and it would go out and work for us.

Now I think how stupid it would be.

DarLorkar

Does not seem to be AI to me.

Just parrots what it sees from others.  Chatbot..yes. AI no.

Steelhelm

How to teach AI about morals, pain, suffering... Good luck

TaishiFox

:O lmao I warned em on FB that this will take over the world!

mgilbrtsn

The thing is, we don't want them full learning capability.  We want them to learn within socially accepted limits. If it learns something outside of it, it will be brought off line to correct the 'fault.'  Socially accepted norms may not reconcile with cold hard facts and statistics.  In these cases, cold hard facts and statistics will be squashed in favor of the consensus view.

A good example of this can be seen in some parts of the muslim world.  We promoted Democracy in places like Egypt, Libya, and Iraq.  They voted, and voted for people and ideas we didn't like.  We then worked to get rid of them in favor of something we really wanted.  What we wanted was a pro western government as opposed to a government that their people actually wanted.  (I'm not making a value judgement on whether policy was good or bad.  That's not my point).  It's the same for computer AIs.  We want something that is different than something we want.

GrumpyHobbit

I posted about this a while back: -

http://forums.mmorpg.com/discussion/448150/twitter-ai-failure-or-not#latest

pingo

After watching Ex Machina @ BURN AI WITH FIRE.

Volgore

We trained chatbots the same way 15 years ago already.
After like 20mins, they were spewing nothing but insults and profanities to everybody.

Good old times when we were rebels!

Quizzical

When computers take over the world and wipe out all humans, they'll only be doing what Internet trolls taught them to do not realizing that they'd actually be able to do it.

Kabulozo

Better to teach a parrot to troll instead.

https://www.youtube.com/watch?v=lUVQz6_-vxc

Agnhar

It just looks like most teenagers angry, immature  and blowing steam on random people on the internet.

Mikeha

Well played Microsoft. 

Malabooga

AI is THE evolution of human kind:) (just needs a bit of tweaking)

Humans are like dinosaurs