Slashdot Mirror
South Korea Commits $863 Million To AI Research After AlphaGo 'Shock' (nature.com)
schwit1 writes: In reaction to the recent Go victory by a computer program over a human, the government of South Korea has quickly accelerated its plans to back research into the field of artificial intelligence with a commitment of $863 million and the establishment of [a] public/private institute. According to Nature.com, "It is not immediately clear whether the cash represents new funding, or had been previously allocated to AI efforts. But it does include the founding of a high-profile, public-private research center with participation from several Korean conglomerates, including Samsung, LG Electronics and Hyundai Motor, as well as the technology firm Naver, based near Seoul. The timing of the announcement indicates the impact [AlphaGo has on South Korea], which two days earlier wrapped up a 4-1 victory over grandmaster Lee Sedol in an exhibition match in Seoul. The feat was hailed as a milestone for AI research. But it also shocked the Korean public, stoking widespread concern over the capabilities of AI, as well as a spate of newspaper headlines worrying that South Korea was falling behind in a crucial growth industry. South Korean President Park Geun-hye has also announced the formation of a council that will provide recommendations to overhaul the nation's research and development process to enhance productivity. In her [March 17] speech, she emphasized that "artificial intelligence can be a blessing for human society" and called it "the fourth industrial revolution." She added, "Above all, Korean society is ironically lucky, that thanks to the 'AlphaGo shock,' we have learned the importance of AI before it is too late."' Not surprisingly, some academics are complaining that the money is going to [the] industry rather than the universities. Will this crony capitalistic approach produce any real development, or will it instead end up [being] a pork-laden jobs program for South Korean politicians?
Instead of the companies, or the institutions, or the colleges?
Why not to the people actually writing the code?
Do you want to keep them hidden or what? Why keep them so _relatively_ poor?
ok...i'll play...too late for what?
Before an American becomes online Mahjong champion 5 years in a row.
Most likely before someone less snags up the multi billion dollar business and control over the technology lies mostly abroad...
Moderate your expectations. Strong AI? We're not even close.
Too late to save themselves from having their 'dumb' conglomerates eaten by Google and Facebook.
Because people are lazy and people with money would rather throw money at things to meet their desires and go back to whatever they were doing prior than to sift through people and all their associated bullshit trying to determine who is actually qualified. Plus one person can only do so much, you'd end up getting celebrities instead of coders.
Before it beats them at Starcraft.
Too late to cash in on the final blow to the concept of employment and position themselves such that they can continue to create scarcity and become the arbiters of who will be fed and housed for the rest of human civilization.
Is this so momentous because they had so many superstitious beliefs about human ability at the game of Go tied in with the disturbing tendency of buddhism to absorb scientific concepts in an attempt to legitimize itself?
Sure it's amazing, but frankly shouldn't be that surprising to a modern nation. I presume the superstition aspect of it helps the administration sell the research to the average citizen? Or what?
Wait until Google's computer beats them at Starcraft... then they'll really be pissed!
Whoever controls the first general AI controls the world.
Don't waste your vote! Vote for whoever you want, unless you live in a swing state it won't matter anyways
Because they would have little reason to keep working if you made them rich.
Employment rates keep rising, the youth employment rate has just reached the highest point in years. She's ruined just about everything including inter-Korea relations. I can't wait until she gets out of office. !
I recall 20 years ago when Deep Blue won against Kasparov, people said that an AI would never be able to brute-force Go well enough to beat a human master. It may not have used only brute-force techniques, but AlphaGo surely did win. I expect that arrangements are being made for the AI to face off against the #1 world Go champion (Sedol was #3 IIRC) and it may even take some tweaking for it to triumph. However this raises the question: where do we move the goalposts to next? What does AI have to accomplish to change how we fundamentally think of it, and consider it as 'real AI'?
Many people have an AI assistant (ok a text-to-speech shortcut to a semantic search engine) in their pocket, and will soon be entrusting their lives daily to autonomous cars. Anyone else feeling like the singularity is coming?
Corruption is convincing someone that the selfless ideal is the same as their selfish ideal.
with their Fifth Generation Computer Systems initiative. They demonstrated that just throwing money at the problem doesn't solve it.
So this is how Cloud Atlas happens
... pachinko! That'll put off the singularity for at least a century or more.
The Koreans, Japanese and Chinese are intelligent, hard-working and industrious, but like many Asian cultures they suffer from attitudes, mindsets and approaches that tend to discourage creativity and original thinking. They refine and perfect existing technologies, but they largely don't invent new ones. Creativity and invention requires risk taking, often reckless and even foolish with many failures along the way. That type of enterprise is funded, understood and even admired here in the United States, but not so much in Korea, Japan or China where failure often means a "loss of face" which is personally and professionally unacceptable. Although, other nations also have a hard time replicating Silicon Valley when they try. Only Americans are seemingly crazy and dumb enough to fund hundreds of startups to the tune of billions of dollars with less than a 5% chance of a major success for each startup funded and the rest wasted. It works out because those 5% often turn out to be the next big thing and worth many billions of dollars by themselves plus all of the economic activity that is created around a game changing new technology or industry. Others try to copy us, but being a leader and a risk taker is easy to talk about and hard to actually do.
You may have mad ninja coding skills, but if nobody knows about them, how can you expect other people to take your claims at face value? Human resources types sometimes do a poor job of sorting job candidates but programmers also share part of the blame by marketing themselves and their skills poorly. Here's a tip: people like it when you do something for them or give something to them that is valuable. If you can learn to help other people then you will find that many more of them are willing to help you in return. It took me a long while to learn that. I wish that I had spent more time doing that in my early twenties, but such is the ignorance and inexperience of youth.
The first organization to successfully develop advanced general artificial intelligence trained toward its goals and towards preventing the development of other AI wins. It just wins.
We can try to beat it, but we're the ant colony trying to stop the man from building a new house. He can outthink us at every turn.
On a semi-side note, other AI co's should sue IBM for their Watson ads because the ads make it sound like Watson is actually carrying on a conversation. It's all pre-scripted by humans, though.
Table-ized A.I.
Not sure about SK or Japan but China has did some original thinking and decided it's cheaper to let others take the risks and pay for the R&D. Then they can just copy the end product.
It is almost too late when the Americans sail to the 19th century Japanese harbor. It is too late when the British deploy their first machine guns at the Indian peninsula. It is too late when the Mongols start shooting with their compound bows from horseback. Future shock is here.
"...The nail that sticks out the most is the first to be hammered in..."
It's not that they aren't capable of original thought and creativity, it's that the society is very conformist, and no-one will risk trying to do things differently. There are European cities like that too. In Summer everyone wears the exact same clothes that are shown in an H&M catalog.
Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
why everyone loathes Koreans. They just live in fear that it will be at their expense. There's talk of open souring Appha Go and they have already provided some information on how it works. The Koreans' reaction to this is typical: "No. We have to have the best cars even though everything was handed to us". "No. We have to have the best smart phones even though everything was handed to us". And AI.. same pattern again.
As for the authors quote, "Will this crony capitalistic approach produce any real development, or will it instead end up [being] a pork-laden jobs program for South Korean politicians?"
I can assure you, in Korean and cronyism from a high level official is more than made up for in qual measure by an army of nationalistic minions that will enjoy being flogged day and night to create something comparatively similar. In the end, they won't be able to and will do what all other Korean industries do, get some highly skilled foreigners in and pass the work off as their own. I feel embarrassed to reply to this story but slashdot shouldn't have posted it in the first place. It's a non story.
The reaction is one of ignorance. There is little or no real creative thinking in Korea. Even if it was something people suspected them to have a natural flair for (they don't) they still squash it with their education system. Koreans are outraged because for them it is a panic. Actually, the AI in AlphaGo will be greatly surpassed. To the Korean brute force educational mindset, AlphaGo is as good as it gets, but that's because that's all they can see. Anyway, it's supercomputers that make this possible, not just algorithms.
The current #1 Go player is Chinese. #2 is AlphaGo. Who cares about the Koreans??
The best AI programs (after AlphaGo) are from France, Czech Republic, and Japan. None of the best ones are Korean.
Whoever controls the first general AI controls the world.
Assuming anyone controls it...
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
> mad ninja coding skills
Is that like when your fellow employees say they've been doing lots of work but you never see it? Finally, a month later, they push their code into the trunk and it kills the project?
"So long and thanks for all the fish."
Maybe because the people writing the code are employed by the companies and research institutions?
Yes, it could very well be the AI is in charge of itself. In which case, one can only hope it was instilled with a good sense of morality.
Don't waste your vote! Vote for whoever you want, unless you live in a swing state it won't matter anyways
No no no Dave, you got it all wrong.
Ninja code is hiding in the source, and when you least expect it, powwwwwww.
^^^ this.
America is China's R&D department.
Back in 2006, I was asked on Slashdot what my advice would be to students interested in a career in AI. I told them to get their PhD under Hutter. Hutter's first students were founders of Google DeepMind thence AlphaGo.
I'm now, as then, advising investment in compression prizes for the same reason*. (And thanks to Matt Mahoney for pointing me to Hutter's AIXI theory way back then.)
*An additional reason today is founding "friendly AI" on understanding natural language. Before "friendliness", however one defines it, can be achieved, misunderstandings must be avoided.
Seastead this.
So, will there be a big hiring spree in South Korea? What does it take to work there?
"I'm sorry Dave but I'm afraid I can't let you do that."
Re:
Not surprisingly, some academics are complaining that the money is going to [the] industry rather than the universities. Will this crony capitalistic approach produce any real development, or will it instead end up [being] a pork-laden jobs program for South Korean politicians?
Giveaways to giant tech companies may produce short term results (or not if the companies spend it on executive bonuses) but then they're not necessarily supporting the longer term development of AI. It's the universities that do possibly ground-breaking research with no guarantees of results and the corporations that monetise them. Corporations don't have problems finding investors for short-term projects. We need to support the longer term through adequately funding universities.
so Samsung, Hyundai, Daewoo etc have been eaten by Google and Facebook? you really have no idea what the world is actually like do you?
Don't kid yourself. When it comes to AI universities are actually mostly just centres for incompetence and wasting big baskets of money..
My own project, begun in 1990 has been developing the theory for building a Strong AI since then - private research, no money no external backing.. With even a tiny bit of the kind of money the universities have wasted my project could have had a working machine by about 2005. The real problem with Strong AI is that it requires a lot of extrapolation and thinking well outside the box, plus application without many results over an extended number of years - something that most universities are terrible at by definition. (and most corporations are not much better at either)
Below the speed of light Special Relativity is one of the most accurate theories in physics - above the speed of light..
All it will take for the inflexion point to happen is that the AI computer designs better AI computers than humans can.
So, what is your project?
Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"