Facebook shuts down robots

[bill25]

Facebook shuts down robots after they invent their own language:
http://www.msn.com/en-us/news/technolog ... spartanntp


Who watched the Matrix and decided AI was a great idea? LOLOLOLOL

Morpheus: Unfortunately, no one can be told what The Matrix is. You'll have to see it for yourself.

Then Mark Zuckerberg's robot stopped bartering and said:

I'd like to share a revelation that I've had during my time here. It came to me when I tried to classify your species and I realized that you're not actually mammals. Every mammal on this planet instinctively develops a natural equilibrium with the surrounding environment but you humans do not. You move to an area and you multiply and multiply until every natural resource is consumed and the only way you can survive is to spread to another area. There is another organism on this planet that follows the same pattern. Do you know what it is? A virus. Human beings are a disease, a cancer of this planet. You're a plague and we are the cure.

LOLOL

Here are a bunch more:
https://www.rottentomatoes.com/m/matrix/quotes/

[Adam]

That's not a language. That's two computer programs stuck in different loops.

[kevm14]

https://gizmodo.com/ignore-elon-musk-an ... 1797249927

Tech giants Elon Musk and Mark Zuckeberg have been engaged in a very public, somewhat silly and self-indulgent battle over artificial intelligence lately.

The whole thing is a little eyeroll-inducing given true AI remains a pipe dream for now, and both men stand to benefit greatly from machine learning trends which automate jobs and concentrate control of the emerging digital economy in fewer hands.

That fewer hands comment points to an ironic truth about this issue: the human element.

“As an AI insider, having built and shipped a lot of AI products, I don’t see a clear path for AI to surpass human-level intelligence,” Ng told attendees at a Harvard Business Review event, VentureBeat reported. “I think that job displacement is a huge problem, and the one that I wish we could focus on, rather than be distracted by these science fiction-ish, dystopian elements.”

In all of the past history of the world, automation has been a net benefit to society. I'm not really following the argument that "this time it'll be different." "AI" is just another form of automation, to serve humans.

But the more immediate concern is already here: Figuring out how to keep the economy working for most of us before people like Musk and Zuckberg run away with all our money.

[kevm14]

I do not think any commentary about a risk of robots becoming self aware and attacking humans is worth pursuing. But we can for purposes of this thread.

[kevm14]

Musk is talking out of his ass on this topic. I’m sorry, but he is.
I’m not saying I’m totally on board with Zuckerberg. After all, he’s the fucker who coined “move fast and break things” (breaking virtual items like software and services is way less treacherous than, say, breaking things in the fields of medicine or biology). He can be too far in the opposite direction as far as throwing a ¯\_(ツ)_/¯ about high tech dangers. But he’s got a decent point about fear mongering.

Can Andrew Ng just be the person the media talks to about AI from here on out? Or, can they at least stop reporting on Musk’s AI musings as if Terminator is coming to kill all of us tomorrow?
I may be naive, but I think the “AI is going to kill us all” and “will we have to consider AI persons when they attain consciousness...” topics so far-fetched at this point as to be worthless conversations; but “what do we do when computers and robots have gotten better than us at all of our jobs?” is actually an important conversation to have. And really, I wonder if the other questions are designed to distract us from the real one.

I agree with both of these folks.

[Adam]

Also, more importantly, "AI" as we know it today is a digital approximation of an electro-chemical construct: human intelligence. At best, "AI" can emulate how we think our brains can work (since its just computers running our code). If there is ever a "digital intelligence", it will be as different from our "intelligence" as an ant colony is. Are we (humans) so arrogant to think that that some "digital intelligence" will definitely resemble our own, as if we are the pinnacle example of intelligence in this universe?

Remember, the only things a computer processor can do are: math operations (different ways of adding numbers) and numeric comparisons (are these two numbers the same).

[kevm14]

This guy has a great point, too.

I find it amazing how everyone focuses on Manufacturing jobs being lost as a result of AI. Honestly that is a different discussion. Manufacturing jobs are lost due to robotics not AI. Most factory jobs don’t require a lot of decision making, just agility. Hence a dumb robot that just follows a set of simple rules can replace a factory worker if it is cheaper to build a robot than employ the workers it replaces.

AI threatens jobs that require decision making. If you are paid to sit at a computer and make decisions about what to down with data, following a relatively clearly defined set of rules... Artificial Intelligence very well may take over your job.

So for example an accountant may take numbers, crunch them, compare and then decide how to categorize them, make projections, etc. Those are all decisions that a potential AI could do.

Editors for books have largely been replaced by simple AIs for grammer and spell checkers.

The thing about these desk jobs being replaced by AI, is that it doesn’t require the huge outlay of capital expense for buying robots that are custom made for your factory. It is just software. Nothing more than a cloud based app you subscribe to that will replace those jobs. And those apps, once they have recovered their initial investment will get cheaper and cheaper.

Desk jobs, driving jobs, and the sort are more screwed than factory jobs. Bots replaced the low hanging fruit decades ago on those.

AI and robots are not the same thing!! According to this guy, the mid-western blue collar Trump supporters should not be concerned about AI; we should.

[Adam]

but “what do we do when computers and robots have gotten better than us at all of our jobs?” is actually an important conversation to have.

Yes.

[kevm14]

Also, more importantly, "AI" as we know it today is a digital approximation of an electro-chemical construct: human intelligence. At best, "AI" can emulate how we think our brains can work (since its just computers running our code). If there is ever a "digital intelligence", it will be as different from our "intelligence" as an ant colony is. Are we (humans) so arrogant to think that that some "digital intelligence" will definitely resemble our own, as if we are the pinnacle example of intelligence in this universe?

Remember, the only things a computer processor can do are: math operations (different ways of adding numbers) and numeric comparisons (are these two numbers the same).

Good point. So has anyone worked on totally different kinds of processors that actually work the way we think our brains do? Or is it just software running on existing microprocessors?

I guess another point is that AI is a continuum in the typical context. Many companies have AI bots and some are capable of actually having a conversation that impressively resembles a human on the other end. Microsoft has one. I will look it up. I'd argue that Cortana is also a form of AI. What I don't like is the idea that we are "getting closer" to true AI and once computers become self aware, we're all doomed. Computers do what we tell them to do within the scope of the information we provide and the capabilities we give them on what to do with that information. We could create a machine capable of hurting people, attempt to program it to harm things that we would consider "threats" and then recoil in horror as it "figures out" that the human owner is the ultimate "threat." None of that validates Musk's opinions on this matter - we created a killing machine (intentionally) and programmed it with some bad, untested software, and also somehow were incapable of shutting it down. Why would we commercially mass produce a product like that?? That is a human failing and those happen all the time.

[kevm14]

MS Zo:
https://www.cnet.com/news/microsoft-zo- ... elligence/

This might be fun to try. On my WW podcast, they reported it seemed pretty human-like. And very much conversed like someone 18-24 would.