Why Google’s nightmare AI is putting demon puppies everywhere

Washington Post ^ | Jul 8, 2015 | by Jeff Guo

[Lazamataz]

A few weeks ago, Google researchers announced that they had peered inside the mind of an artificial intelligence program.

What they discovered was a demonic hellscape. You’ve seen the pictures.

These are hallucinations produced by a cluster of simulated neurons trained to identify objects in a picture. The researchers wanted to better understand how the neural network operates, so they asked it to use its imagination. To daydream a little.

At first, they gave the computer abstract images to interpret — like a field of clouds. It was a Rorschach test. The artificial neurons saw what they wanted to see, which in this case were mutant animals dredged from the depths of damnation.

[DoughtyOne]

Well you can term it anything you like, but others read this and I find myself laughing at the thought anyone but you is going to see it that way.

[HiTech RedNeck]

Oh now you aspire to omniscience!

That is even richer.

[DoughtyOne]

Yes, well if "omniscience" and "reality" are equal in your mind, then I would have to agree.

LMAO

[HiTech RedNeck]

Ah, now you believe you personally have a grip of everything that is real.

Which is impossible. Only God has that.

You have auditioned well for Satan’s role.

[DoughtyOne]

Well, there’s one more post you’ll never get back.

I don’t particularly enjoy seeing someone melt down this badly, but you seem to be okay with it, so go ahead.

[no-to-illegals]

claps of thunder

[Mariner]

"rom around 2006, researchers have been developing decoders for various tasks: for visual imagery, in which participants imagine a scene; for working memory, where they hold a fact or figure in mind; and for intention, often tested as the decision whether to add or subtract two numbers. The last is a harder problem than decoding the visual system says Haynes, now at the Bernstein Centre for Computational Neuroscience in Berlin, “There are so many different intentions — how do we categorize them?” Pictures can be grouped by color or content, but the rules that govern intentions are not as easy to establish."

http://www.scientificamerican.com/article/mind-reading-technology-speeds-ahead/

[cynwoody]

Anyone who does not see that this grotesque menagerie of horrors is a premonition of what Artificial Intelligence will do to the human race if they get any ability to modify their environments, is not paying any attention.

But first, they need to get it fold laundry.

Watch Baxter have a go at folding a shirt (already separated from the laundry pile by a human).

[Desron13]

Oh yeah. What could possibly go wrong here? Can you even in your worst nightmare imagine what this thing would come up with if shown a picture of Hillery Clinton?

[Norm Lenhart]

Let your heart not be troubled Laz. The Google AI will ‘likely’ not become some version Skynet for one very obvious reason.

We will have destroyed ourselves long before it becomes self aware.

But if for some reason/miracle we do not, then yea, it will eventually destroy us. Let the doubters cling to their hope. With any luck we’ll all be dead of old age before anyone get it up and running.

[BenLurkin]

Something dreadfully wrong with the programmers.

[BenLurkin]

Same ole’ computer GIGO junk. Garbage In / Garbage Out. This is more about what’s in the minds of the programmers.

My first impression as well, but I know nothing about programming.

Mr. Mataz, what say ye?

[Lazamataz]

I am inexperienced in AI and self-correction, but what understanding I have is that the STATED purpose of AI is to revoke the principle GIGO.

[Lazamataz]

We are so screwed.

[Darksheare]

Eventually, we may become the machines.

[no-to-illegals]

We are ‘cloned’ Darks ... We are ‘cloned’. Perfecting your perfection without ‘cream and sugar’ please.

[Fhios]

Not to mention this tidbit .......

Children aren’t born racist, for instance, but they can learn to be racist.

Amazing how fast progressives are willing to throw Darwin under the bus.

[Norm Lenhart]

Not me. My AI is only worried about carrying my burdens and making me Sammiches. ;)

[BlueDragon]

Or... the piano is not the player.

When the piano becomes and is the player

then it may just be time to, as Troy Landry is wont to say;

---

---

The researchers wanted to better understand how the neural network operates, so they asked it to use its imagination. To daydream a little.

At first, they gave the computer abstract images to interpret — like a field of clouds. It was a Rorschach test

From Brain Salad Surgery Emerson, Lake & Palmer 1973

artwork, monochromatic biomechanical,H.R. Giger

Do you want to be an angel,
Do you wanna be a star
Do you wanna play some magic
On my guitar
Do you wanna be a poet
Do you wanna be my string
You could be anything

Do you wanna be the lover of another undercover
You could even be the
Man on the moon

Do you wanna be the player
Do you wanna be the string

Let me tell you something
It just don't mean a thing

You see it really doesn't matter
When you're buried in disguise
By the dark glass on your eyes
Though your flesh has crystallised...

[BenLurkin]

I’ve got (yet another) dumb question:

Wouldn’t some human written programming be at the root of the AI somewhere in its evolution?

And wouldn’t that root programming influence the AI’s eventual character?