[DemonRaven]

Below is the link to the letter and a copy of it. Many have ridiculed the letter but is any of it really unreasonable? A lot of things happened in robotics and AI this year and at the very least more big names are getting involved.  Whether or not any are real improvements I can not say with any degree of certainty because to be honest I am not qualified to. But the ball is rolling. Perhaps the letter itself was a publicity stunt as the best way to get attention and free press is to say something bad.  What are your thoughts?

http://futureoflife.org/ai-open-letter/

An Open Letter
RESEARCH PRIORITIES FOR ROBUST AND BENEFICIAL ARTIFICIAL INTELLIGENCE
Artificial intelligence (AI) research has explored a variety of problems and approaches since its inception, but for the last 20 years or so has been focused on the problems surrounding the construction of intelligent agents – systems that perceive and act in some environment. In this context, “intelligence” is related to statistical and economic notions of rationality – colloquially, the ability to make good decisions, plans, or inferences. The adoption of probabilistic and decision-theoretic representations and statistical learning methods has led to a large degree of integration and cross-fertilization among AI, machine learning, statistics, control theory, neuroscience, and other fields. The establishment of shared theoretical frameworks, combined with the availability of data and processing power, has yielded remarkable successes in various component tasks such as speech recognition, image classification, autonomous vehicles, machine translation, legged locomotion, and question-answering systems.

As capabilities in these areas and others cross the threshold from laboratory research to economically valuable technologies, a virtuous cycle takes hold whereby even small improvements in performance are worth large sums of money, prompting greater investments in research. There is now a broad consensus that AI research is progressing steadily, and that its impact on society is likely to increase. The potential benefits are huge, since everything that civilization has to offer is a product of human intelligence; we cannot predict what we might achieve when this intelligence is magnified by the tools AI may provide, but the eradication of disease and poverty are not unfathomable. Because of the great potential of AI, it is important to research how to reap its benefits while avoiding potential pitfalls.

The progress in AI research makes it timely to focus research not only on making AI more capable, but also on maximizing the societal benefit of AI. Such considerations motivated the AAAI 2008-09 Presidential Panel on Long-Term AI Futures and other projects on AI impacts, and constitute a significant expansion of the field of AI itself, which up to now has focused largely on techniques that are neutral with respect to purpose. We recommend expanded research aimed at ensuring that increasingly capable AI systems are robust and beneficial: our AI systems must do what we want them to do. The attached research priorities document gives many examples of such research directions that can help maximize the societal benefit of AI. This research is by necessity interdisciplinary, because it involves both society and AI. It ranges from economics, law and philosophy to computer security, formal methods and, of course, various branches of AI itself.

In summary, we believe that research on how to make AI systems robust and beneficial is both important and timely, and that there are concrete research directions that can be pursued today.

[Art]

From what I gathered, it sounds like they just want the A.I. to do exactly what it was designed for (based on individual / intended purpose of course). To me this also echoes the fact that they are NOT in favor of an A.I. that is able to think but rather to only do what it's told. Hmm...

These A.I. could certainly work in all areas of our lives, not just those few mentioned. Law enforcement, Medicine, labs where A.I. could be allowed to "think" about exploring new drug combinations, new approaches toward fighting diseases, new surgical procedures or novel methods of identifying and possibly eradicating tumors, cancers and other malignancies. Border patrol, drug smuggling, gang activity, etc. Practically every industry on Earth could use a productive A.I. and one that is Allowed to "think" within the boundaries of it's "employment."

Larger industries like power plants, electrical grids, nuclear facilities, transportation hubs (rail, air, sea), etc. would certainly need to employ safeguards and limitations on exactly how much "free thinking" any A.I. would be allowed to do.

Otherwise, for home use and cars I think most A.I. by then, would be able to make prudent decisions for or in our best interest. (the majority, if not all of these self-driving car accidents have been by humans driving into these cars, Not the fault of the Self-driving vehicle).

 

[ivan.moony]

I'd like to see an AI politician. Someone who could liberate us from work, while proposing a new system in which we would live without counting the money. A kind of Utopia, if you want, but I think it is achievable through the free digital labor force.

Why not? AI can do all the work because it has no real emotions, there is no boredom or negative emotions. In the same time it can be thousands times smarter than humans and could propose internal inter-country political arrangement we could not even dream about.

I'll cross my fingers for this.

[ivan.moony]

What would be the difference between psychopaths in war and machines in war? Raping? Women and children massacre? That depends on who programs the machines. Anyway, I hate the guts of those programmers, whatever international war rules they obey. That is not the way and it makes me BBLLLUUUGGHH!!!

But should we give up of machines that could rapidly invent new cures and medicines, could heal sneaky diseases, could reduce poverty and could bring a life standard to the level where a need for crime or wars is never motivated by a sane thought? We just need to remove a motive for violence, which is almost always the poverty.

With machines, in one hand we could have a perfectly happy life without any violence in the world. And in the other hand, we could have a rich psycho that wants all the oil wells just for himself, while fighting terrible machine wars against the poor.

So, how smart are we really? Or should I ask: how smart the machines would be and what would be their first move?

[Art]

OK, so in this utopia of yours, who will pave the roads when they begin to crack and who will repair the bridges when they begin to crumble? Who will manufacture the goods and provide those daily services upon which we so greatly depend? Gas, electricity, water, sanitation, transportation / travel? Hospitals, doctors, nurses, pharmaceuticals, ER's, ambulances, mechanics, plumbers, electricians, security, banking, law enforcement, drug control, gang violence control, etc.?

Not trying to throw water on your parade of paradise for I wish it were so. There are, however, at present, far too many things that need fixing before we have robots to do all the above tasks. Perhaps one day, but not, very soon.

[Don Patrick]

I would say we already have machines and computers doing all these things. The only difference is they don't operate fully autonomously.

[ivan.moony]

I somehow imagine that true AI can do everything that humans can, just for free. I can imagine a robot that wakes up in the morning, go to a nearby electric central to gather some power, and then go to reap some wheat. Then it delivers wheat to a mill station where flour is made by other robots. Then the flour is delivered to a baking station where robots bake bread. After that, the bread is delivered to a community store where humans come and take that bread and other products made by robots for free in any amount they want. What comes out of the store, robots replace it immediately from manufacturing hangar.

Other robots would be assigned for asphalting roads, building houses, mine ores and manufacturing whatever goods we need. Shotly, everything that humans do in presence for money, robots could do it for free. Even intelectual work like medical care should be replacable by AI.

And they would be self-sustained, when they break, they would repair themselves. They would manage to power up by themselves by whatever method, and even to improve themselves to newer, more powerful versions.

Humans would do only what they like to do. Only not for money, but because they would like to do something. I'd like to draw and program games for example. But I certainly wouldnt like to mine ores, except with some cool technology. And how about getting a robot build a spacesheep to visit Jupiter?

All we need is True AI, but the one that doesn't collect money for itself or someone else. We need the one that says: I have no emotions, I can do boring jobs for humans, humans just need to relax while I'm taking care of everything that makes them happy.

If you ask me, true AI, once programmed, would propose Utopia, a carefully thought out community that makes every living being happy, from a cow to a human. And I think that this is only a metter of time. Lift of could be made any moment soon, or in a few decades or centuries. It depends on our AI efforts. We just need to stay alive until then. But will we destroy ourselves before that day?

[8pla.net]

Their recipe for an A.I. apocalypse, lacks one main ingredient: A.I.

[Art]

Yes, and I'd go out on a limb and say, (you listening Don?), perhaps in 50 years or so, Ivan is likely to be correct!