Scientific American article on Computer Chess

[Lyudmil Tsvetkov]

it is an electrical circuit immediately connected to the body, and able to assess events as positive or negative. able to feel pain and joy.
machines are not able to do that.

Says who? The 'Tsvetkov Oracle'?

Certainly the body is nothing special here. If we succeed in connecting a human brain to a mechanical body (or just enough to let the brain communicate with the world) then surely we still have "real intelligence".

And is there anything special about the brain that could not be simulated artificially? That seems extremely unlikely.

no, the basic precondition for intelligence to arise is to have a living soul.

no living soul, no intelligence.

adaptation is the key to intelligence. no adaptation, no intelligence.

humans are able to adapt, machines not. in order to adapt, one needs to feel either comfortable or incomfortable. only the human or animal body, a living soul, can feel comfort and discomfort. not machines.

if nothing bothers you at all, no need to adapt, no need to change, no need to understand/intelligise. once your body feels pain, you need to change something about it; if a condition is comfortable, you would like to have more of it. people built shelters, because they felt cold. and stuck to fire, because fire brings warmth.

no adaptive brain without a living soul.

computers are just and will remain for ever just machines, maybe severely outcalculating humans, but still unable to adapt.

basic computers always existed. for example, if you want to measure which end of a beam is heavier than the other, you take a steady point, put the beam across, and, due to gravity, the beam will sink towards the heavier end. the steady point-beam computer has already calculated a thing. is this primitive computer intelligent?

similarly, when you try to calculate how much 34567 times 87654 makes, without machine and pencil aid, you might be able to do that in a time, maybe not, but certainly not, if someone puts a very hot object to burn your skin. then you will simply not be able to do anything. not only that, but suffer. and all the while, you have been solving the very same task.

of course, how much the multiplication of 2 big numbers makes does not matter at all, all that matters is that you are happy you have been able to compute that task quickly and painlessly, struggled a lot with it, or, even instead of solving it, having burnt your skin.

[syzygy]

no, the basic precondition for intelligence to arise is to have a living soul.

Feel free to repeat yourself, but without actual arguments from your side there is no need to address anything.

[Rein Halbersma]

Artificial intelligence (AI) is intelligence exhibited by machines. In computer science, the field of AI research defines itself as the study of "intelligent agents": any device that perceives its environment and takes actions that maximize its chance of success at some goal.[1]

[1] The intelligent agent paradigm:

• Russell & Norvig 2003, pp. 27, 32–58, 968–972

The definition used in this article, in terms of goals, actions, perception and environment, is due to Russell & Norvig (2003). Other definitions also include knowledge and learning as additional criteria.

Certainly a chess engine perceives its environment (the chess board) and takes actions (selects moves) that maximizes its chance of success at some goal (that of winning the game). Chess engines also encode knowledge, and many include a basic form of learning.

So a chess engine is an example of AI as that term is used in the field.

Whether there is "real intelligence" in a chess engine is another question. To answer it one first needs to have a definition of "real intelligence".

If a chess engine does not exhibit "real intelligence" because we understand how they work, then I guess the human brain will stop being referred to as "intelligent" once we fully understand how it functions...

Indeed, according to Figure 2.6 of the cited Norvig & Russell book, a chess program operates in a highly strucured environment, that is Fully observable, Multi-agent, Deterministic, Sequential, Semi-dynamic and Discrete.

Go falls in the same category, and despite the huge recent successes of AlphaGo and its deep learning components, computers are still a long way away from artificial general intelligence (i.e. high performance over the whole range of far less structured environments in which humans have to operate). See e.g. this blog: https://medium.com/@karpathy/alphago-in ... 7718cb95a5

[syzygy]

Indeed, according to Figure 2.6 of the cited Norvig & Russell book, a chess program operates in a highly strucured environment, that is Fully observable, Multi-agent, Deterministic, Sequential, Semi-dynamic and Discrete.

Go falls in the same category, and despite the huge recent successes of AlphaGo and its deep learning components, computers are still a long way away from artificial general intelligence (i.e. high performance over the whole range of far less structured environments in which humans have to operate). See e.g. this blog: https://medium.com/@karpathy/alphago-in ... 7718cb95a5

We don't know how to create artificial general intelligence, which probably is the ultimate goal of the AI field. We may even feel we have not come one step closer to achieving that goal since the days of Turing.

But at the same time computers are now becoming pretty good at a range of "intellectual tasks" that seemed out of reach not that long ago. Google Translate may not be perfect, but I can use it to understand texts in languages that I know nothing about. So what if it is only "counting words" and not "understanding" anything? Computers are there to serve us and they are serving us better and better.

[Cardoso]

The problem is that humans tend to exagerate a lot and put a lot of faith on science (wich is putting a lot of faith on other humans), making promises and predictions way way beyhond their capabilities.
Other areas where humans promises a lot is politics.
Another one is space exploration, there are scientists that take for granted and as a natural thing for humans to do, space exploration. Well some day we might accompplish that. But not for the near future, for now we just keep orbiting Earth, and that isn't space exploration. We don't even have space ships, we have something more in the likes of space rafts. For example I believe we could go to Mars (as if that goal is really important), but with current technology we could do it just barelly or perhaps even not possible. And Mars is just our neighbouring planet (the moon doesn't count, as it is a satelite or a dual system whatever).
A real space ship is something that allow us to get to Saturn and when we get there we can say: oh I forgot my camera, and turn back to Earth to fetch the camera, and resume the mission in Saturn with 99.99% fuel autonomy. This is just one of many aspects of a real space ship. What about our neighbouring star Alpha Centaury at 4 light years distance? Forget it. In space the distances are absolutely astonishing/terrifying.
So in short people usually promises a lot, but delivers too little. I've seen very highly educated scientists saying things that are more proper of 13 year old kids. But I guess that's what humans are.

Does anyone still remember Intel's promise in the year 2000? They said in 2010 they would have 10GHz cpu (yes I know we are just just getting there). What about that THz (TeraHertz) transistor they said they had quite a few years ago? I wonder if those articles are still on the internet, since they are from many years ago.

[syzygy]

The problem is that humans tend to exagerate a lot and put a lot of faith on science (wich is putting a lot of faith on other humans), making promises and predictions way way beyhond their capabilities.

I'm not really sure what the "problem" would be.

It is true that in the past AI academics have promised too much, I suppose in an attempt to attract more funding. But if IBM and Google spend money on AI, then that is not to attract funding but to produce something that works and that, ultimately, they can sell. (Of course that does not exclude the possibility of failure.)

The ones who are making the promises (or rather, threats) are still the academics, this time people like Hawkins who just need the attention.

Another one is space exploration, there are scientists that take for granted and as a natural thing for humans to do, space exploration. Well some day we might accompplish that. But not for the near future, for now we just keep orbiting Earth, and that isn't space exploration. We don't even have space ships, we have something more in the likes of space rafts. For example I believe we could go to Mars (as if that goal is really important), but with current technology we could do it just barelly or perhaps even not possible. And Mars is just our neighbouring planet (the moon doesn't count, as it is a satelite or a dual system whatever).

All such things will go step by step .You can't go to Saturn before mastering Mars. You can't go to Mars before mastering the moon. The spaceship that gets us to Mars will probably not suffice for Saturn, but every step teaches us new things without with which we could not make the next step.

After the moon landing, Mars would have been possible in the 80s or even earlier, but it would have cost an awful lot of money and there was not even the political will to finish the Apollo program. But technology has been developing in other ways, so not all the time that has passed is lost.

I've seen very highly educated scientists saying things that are more proper of 13 year old kids. But I guess that's what humans are.

I haven't seen many scientists predicting a visit to Alpha Centauri within our lifetime. There have been scientists predicting a trip to Mars, but scientists are not politicians.

Does anyone still remember Intel's promise in the year 2000? They said in 2010 they would have 10GHz cpu (yes I know we are just just getting there). What about that THz (TeraHertz) transistor they said they had quite a few years ago? I wonder if those articles are still on the internet, since they are from many years ago.

The prediction was 10ghz by the year 2011. Intel was laughed at, because a child could calculate that computers would be running at 128ghz in 2011.

https://www.geek.com/chips/intel-predic ... 11-564808/
See the original user comments and see those from 2011.

But to turn back to the Scientific American article, does it contain any unreasonable prediction? What is there to criticise in it?

[duncan]

What about our neighbouring star Alpha Centaury at 4 light years distance? Forget it. In space the distances are absolutely astonishing/terrifying.

the next best thing is sending a probe to Alpha Centaury.
it is not impossible that this will be sent within 25 years

https://www.scientificamerican.com/arti ... est-star1/

For Yuri Milner, the Russian Internet entrepreneur and billionaire philanthropist who funds the world’s richest science prizes and searches for extraterrestrial intelligence, the sky is not the limit—and neither is the solar system. Flanked by physicist Stephen Hawking and other high-profile supporters today in New York, Milner announced his most ambitious investment yet: $100 million toward a research program to send robotic probes to nearby stars within a generation.

“The human story is one of great leaps,” Milner said in a statement released shortly before the announcement. “55 years ago today, Yuri Gagarin became the first human in space. Today, we are preparing for the next great leap—to the stars.”

“Breakthrough Starshot,” the program Milner is backing, intends to squeeze all the key components of a robotic probe—cameras, sensors, maneuvering thrusters and communications equipment—into tiny gram-scale “nanocraft.” These would be small enough to boost to enormous speeds using other technology the program plans to help develop, including a ground-based kilometer-scale laser array capable of beaming 100-gigawatt laser pulses through the atmosphere for a few minutes at a time, and atoms-thin, meter-wide “light sails” to ride those beams to other stars. Each pinging photon of light would impart a slight momentum to the sail and its cargo; in the microgravity vacuum of space, the torrent of photons unleashed by a gigawatt-class laser would rapidly push a nanocraft to relativistic speeds.

[Dirt]

the next best thing is sending a probe to Alpha Centauri.
it is not impossible that this will be sent within 25 years

Not impossible, but it's not going to happen. It's not that important to us. Maybe we'll try it in a century but it will probably take much longer.

Getting humans to another star will take at least millennia, but in the meantime we can learn much with giant space based telescopes. If we can find convincing signs of extrasolar life relatively nearby it would accelerate things somewhat.

[duncan]

the next best thing is sending a probe to Alpha Centauri.
it is not impossible that this will be sent within 25 years

Not impossible, but it's not going to happen. It's not that important to us. Maybe we'll try it in a century but it will probably take much longer.

Getting humans to another star will take at least millennia, but in the meantime we can learn much with giant space based telescopes. If we can find convincing signs of extrasolar life relatively nearby it would accelerate things somewhat.

its important to milner who is funding it and if he solves the problems and shows it is feasible, then it just becomes a political issue do we want to pay for photos of planets of a star.

[MikeB]

https://www.scientificamerican.com/arti ... ing-chess/

The article was not worthy of this many posts as it came up pretty short to Scientific American standards. It was written for the average person who knows almost nothing about AI or anything about computer chess and was clueless about AI 20 years ago. The definition of AI has change over last 20 years - back then chess engines were considered AI, now they are considered pretty far removed from AI , but then one must crawl before you can walk, so it was stepping stone that had to be hurdled. The thing about AI, you need umderstand why? It just cannot be because the black box says so. And we're still quite aways from that.