Hacker News new | past | comments | ask | show | jobs | submit login
China’s SenseTime, the world’s highest-valued AI startup, closes $620M round (techcrunch.com)
184 points by alex_young 10 months ago | hide | past | web | favorite | 113 comments

While this article notes that the Chinese government is one of Sensetime's most visible partners, it doesn't really make it clear the extent to which private sector companies like SenseTime are inseparable from the public sector in China. By supporting SenseTime as its client, Beijing has signaled to investors like Alibaba that it has chosen a winner, and that it will never let SenseTime fail. As with most mainland tech companies, SenseTime will heed the diktats of the central government, much as social platforms like Weibo march in step on censorship requests.

One interesting question this raises is: How can non-Chinese compete against such national champions, this intense cooperation between public and private sectors? In the US, those relationships are much more problematic, as the recent uproar at Google about Project Maven illustrates. Not only that, but the US government is neither as aggressive nor as forward-looking about such partnerships, preferring to let market forces have their way. China is tilting the playing field in its favor, while the world's sole superpower is hobbled by clowns and criminals in the executive branch, and corrupt prime contractors pretend to supply the government with tech that they neither understand nor make themselves.

When other countries get brought up, if it doesn't relate to something the US is a leader in, half the time the subject just gets undermined or discredited somehow. Not just China, but even with other Asian/African/European/South American countries. There's some mentality of there's America, and then there's everyone else in the world. I'm sure not everyone has this view, but it comes off as pretty damn self-righteous. It seems like global matters are treated with a double-standard where negatives about America get treated with "C'est la vie", but then other countries people haven't even visited get judged like some "lesser-than".

For example, coming back to the US from studying in Germany, people asked what the conditions were like and it was pretty interesting there was this imagination of migrants running wild, riots in the streets, like it was 1939 again when in reality things were pretty much the same. Unsurprisingly, I'm sure the US media doesn't help with their approach to sensationalism.

What specific negatives about America get 'treated with "C'est la vie"'? Americans and the American media are extremely self critical.

To counter your anecdote, when I talk to many Europeans there is this imagination that American cities now have gun-wielding criminals "running wild, riots in the streets" when in reality things "were pretty much the same".

People who don't know much about places outside their home often fall victim to sensationalist media. And Americans tend to be very inward focused so they are vulnerable to this. However, to say Americans ignore their own problems is blatantly false.

Are you kidding me? Don't get me wrong I love america for being the epicenter of western culture and its innovation - I even plan on living there for some years. I feel alienated by my own German culture with its weird lederhosen, mozart and the general sentiment for illiteracy regarding all things digital.

As far as I know americans think "God" created (or blessed or whatever) their holy country and its the greatest nation on earth. There is the classic american patriotism. The pledge of honor (and that kneeling thing??). Almost half of americans (or at least a very large Marjory) are creationist and believe angels are real. Even some of your presidents do. There are actual fights weither or not evolution should be taught in schools. Your relationship with Religion is also borderline pathological. There are studies that show higher developed countries tend to have atheist or at least agnostic leanings but the US is an outlier here.

Your irrational war on drugs is a disaster, it, together with your racist history of slavery and segregation causes your cultural critical problems with your minority communities. Your police acts like there in a war zone. Your media is a shit show of of people throwing mud at each other. If you ever learn german you should watch the Tagesthemen which is the most unbiased and informative news I have ever seen in my life. Alternately I think french24 comes close though I'm not sure.

Your relationship with guns is a whole other story as your own students can attest. I can understand the attraction to guns but Jesus you behave like an alcohol addict saying its only medicine.

Some other worthy mentions are your commercial prison system that gives you the highest incarceration rate worldwide. I also heard many times your Judaical system highly disfavors the financially weak but I don't know much about that topic. You also have a "fat" epidemic as I have heard. Also your education system is kinda bad and too expensive.

To close it off - your politics are highly polarized with borderline retarded conservatives that call everyone soyboys on the one side and don't even get me started on the social justice warriors that took a University president hostage[1] and in general know nothing else but to scream loudly.

Cooperation oligopolies are gaining too much control and you see anything remotely welfare related such as the obama health care or free universities as communism.

Btw. education is gonna be the most important field in the future. With rising demand for high skilled jobs poor people are not gonna have it easy. So free universities aren't that bad of an idea (and yes obviously they are not free but paid for by the tax payer).

Also german A-Level History Lessons are of extremely high qualities so we know all about our past and what lessons that brings with it. One of the most important skill I learned in School was in History where we learned to critically analyse any kind of source. It kinda makes you immune to bullshit media of all sorts.

EDIT: I didn't read the source, it was simply the first google match and I just put it there for reference.

1 https://www.zerohedge.com/news/2017-05-26/they-believe-i-was...

I find it hilarious that you praise your own critical thinking skills while citing the first Google result that you didn't read. You sound like one of these hypothetical Americans that never criticize themselves! The biggest problem in your country is lederhosen and Mozart in your opinion? And they say Germans can't be funny...

It makes sense to me that you didn't read your source though, you didn't read my post either. I didn't say America doesn't have problems, we have lots of them! I just said that we are aware of them, and often criticize them. For example, you just listed off a bunch of grievances that the majority of New Yorkers would agree with. (I currently live in New York and the people here are also somewhat disconnected with the realities of middle America)

I really like and respect Germany, and I especially agree with you that the news there is better than in the US. But isn't that the point? Many Europeans watch American news and get sensationalized opinions of what is going on here. It is also worth noting that your country is extremely homogeneous by American standards (partially due to your own country's racist history), and you seem to be assuming America has one unified opinion when that is not the case. We are constantly arguing with ourselves on a much wider range of issues than what Germany must consider.

You are clearly not immune to bullshit media, because you have just exaggerated every stereotype that Europeans have about Americans and stated it as a fact.

Not judging your other points, but if you say Tagsschau is the most unbiased program, then you have read too much Der Spiegel.

I did not say that for the record

I'm not American but my god you have things so wrong. How long have you lived in the US?

You say you're immune to bullshit media, yet you link zerohedge, and your entire view of things here (at least culturally) seems to be based off of sensationalist media and things said on reddit etc. I've never heard the term 'soyboy' in real life, most religious people are pretty nice, in fact most people aren't even religious to begin with, most conservatives are reasonable, 99.99% of gun owners are responsible.

I think you're wrong about America's 'relationship with religion' being pathological. In fact, Utah (the most religious state probably, and pretty advanced) has almost no homelessness. Contrast that with the most liberal/atheist states and it's pretty much the opposite. So, the church can play a positive role. That said, religion isn't actually relevant in many parts of modern US society.

As for guns, I think its a mental health epidemic with a bunch of columbine copycats, rather than a strictly gun issue. I could say that Europe has bad relationship with trucks.

The war on drugs is a disaster yes. The prison system is a disaster. The fat epidemic is a disaster (I blame corn syrup aka corn subsidies and car culture), but there are some pretty dang fat people in Europe too. Education quality depends on the state, but the northeast (MA, CT, NJ) has some great public schools. The political divide is a disaster. Healthcare "policy" is a disaster.

The US is so large and diverse that states act more like european countries, each with their own stereotypes and policies. For example, gun culture is non-existant in the northeast, yet in the west and south it is very prevalent. I mean, in NYC carrying even a folding knife is illegal, yet you're probably considered weird if you don't have one on you at all times in other parts of the country. In the EU, you have Greece and Switzerland, but they couldn't be more far apart culturally and even financially.

Europe does a lot of things better than the US, especially with infrastructure (new infrastructure - our old infrastructure is awesome we just haven't been investing in it for the last 30 years due to the small government conservitard meme). Overall QOL is probably higher.

As for what you do worse, a lot of EU states are nanny states with very high taxes. It doesn't seem like there's a lot of upward mobility. Salaries are low, but there is still inequality. You don't pioneer much.

Just as a fun aside, the 'housing first' model that UTah adopted...was done first by LA and NY back in the late 80s/early 90s.

Also, Utah has plenty of homeless. What they've reduced (to 1/10th of what it was) is -chronic- homelessness. They still have 14k people per year spending one or more nights in an emergency shelter each year ( https://jobs.utah.gov/housing/scso/documents/homelessness201... , page #35), which is essentially unchanged since before the 'housing first' model was adopted.

What is especially interesting is that due to the inter-state differences...raw homeless numbers aren't really that telling. That is, places that are friendly to the homeless tend to have more homeless. Which...makes sense. Even Utah despite its housing policy, is not considered particularly friendly to homeless people, and there is some evidence that people leave the state when homeless ( https://www.theguardian.com/us-news/2017/sep/22/salt-lake-ci... ).

I basically mention this to point out that UTah isn't quite the shining beacon on how to 'solve' the homeless situation as they were touted a few years ago..and that at best their religiosity is irrelevant to that.

> our old infrastructure is awesome we just haven't been investing in it for the last 30 years due to the small government conservitard meme

I agree with the majority of your comment, but the reality of this part is actually considerably more complicated. I used to think the same way, but it turns out that we are incredibly horribly bad at efficiently doing infrastructure projects, with costs/mi tending to be some multiple of similar projects in other developed countries. This holds even when controlling for all sorts of things; my understanding is that economists in the relevant field aren't really 100% sure what the cause is.

From this perspective, the small govt argument becomes a lot more grounded in real complaints about govt's incompetency at this task, on top of their baseline ideological complaint about government not being an appropriate way to allocate capital in the first place.

FWIW, I'm pretty irritated at the horrendous inefficiency of American infrastructure projects but also think that they often are economically worthwhile despite it. But it's unfair to cast this complaint as a thoughtless "conservitard meme" in the case of infrastructure, given that the complaints are to some degree grounded in reality.

>As for guns, I think its a mental health epidemic with a bunch of columbine copycats, rather than a strictly gun issue.

Which is precisely the issues most( not just EU ) don't understand America.

It's because of the Pax Americana brought forth by America's global economic, political and cultural dominance since WWII.


Globalization is a one way street with the implicit assumption that we are at the top of the food chain and make all the progress, innovation and rules.

The 'insignificant other' crashing this party leads to all sorts of dissonance visible in every comment thread on any kind of progress in non-western countries.

This is intrinsically tied to identity so when people say we are in a global world its a very superficial construct hiding some dangerous truths for the naive and uninformed.

Like any good futurist or cyberpunk novel, the lines between "government," "corporation," and "individual" are blurring.

Imagine how much power Comcast would have if it silently went rogue and threatened to expose the pornographic habits of a few key senators? Or if shell, Chevron, and BP formed a cabal and declared an oil embargo on the USA, refusing to dock their tankers until they were given tax breaks or until the USA put tariffs on competition.

Right now China is reminding the world that governments still are the sovereign powers. I wonder when the scales will tip?

The executives that made those decisions would be found, arrested, and imprisoned, if they were lucky. You can only jerk the little guy around like that. Knowledge is not power, power is power.

> The executives that made those decisions would be found, arrested, and imprisoned, if they were lucky

To give mechanism to this method, in America power is competitively shared. Those the ISP is blackmailing will stay quiet. But there are others intensely motivated to uncover the crime, both inside the blackmailed politicians' orbits (to preserve their careers post-downfall) and outside it (to boost their careers by taking out a competitor, to themselves or their boss).

Knowledge is power-lite; when properly wielded, you can convert it into reservoirs of actual power.

The world has always had blurred lines between government corporation and individual.

There are too many examples to name but the big ones are the Dutch East India Company, Nationalized Railroads in the US in WWI and more recently the "Bailout" of banks in the wake of the real estate collapse.

What we've yet to see in history is a corporation that directly and openly challenges a state for sovereignty. I've written in the past that it would be technically possible for a large corporation to raise an army/navy,buy land and declare sovereignty, but that none deign to do it.

The Catholic church is basically a corporation, where the cardinals (the board) elect a non-hereditary emperor (the CEO). Theories of the firm have modeled ideas about corporate governance on the success of this structure. In the past this catholic firm raised armies, held large areas of Italy as the papal states and still today are considered to be a sovereign entity.

> What we've yet to see in history is a corporation that directly and openly challenges a state for sovereignty.

I think you're correct, but the East India Company came pretty close, didn't they?



As I mentioned in my statement, they certainly come the closest but still far from it, given that they were chartered by the King and then their charter revoked as soon as they grew too large with their land holdings (India) effectively confiscated.

They also didn't challenge Indian state directly, but rather as vassels of the British Empire.

What I'm talking about is for example if Amazon raised a Military, set up a border around Seattle, declared the Port their own and then told the US Govt "we're a sovereign nation" and forced the US into a treaty.

Don't think an army assembled in this manner could withstand the US military, at least not yet

Unquestionably not.

I can think of one example of a corporation overthrowing a state, albeit with its parent government's assistance: https://en.wikipedia.org/wiki/United_Fruit_Company

Still falls into the same category of being given a state granted monopoly.

At the point that a corporation has an independent military then we might see something interesting. I'd argue they would get obliterated fairly quickly but it would be an interesting effort.

Why make it overt if you can just buy the government?

Unrelated but for those wondering, "diktat" is not a misspelling of "dictates" as I initially thought. Instead, "diktat" means "an order or decree imposed by someone in power without popular consent" whereas "dictates" means "an order or principle that must be obeyed".

Think "the dictates of the fashion community" versus "The diktats of the dictator"

> Not only that, but the US government is neither as aggressive nor as forward-looking about such partnerships

What? The chinese modeled their economic system after the US. We love to portray ourselves as having a separation between the "private" and the "public". That is absolute nonsense.

The chinese, like the US and europe and japan and south korea and all major world economies, are corporate mercantilists. When the railroads, mining corps, oil corporations, etc needed the natives to be "gone", it was the US government that exterminated the natives. When US corporate interests needed access to central america, it was the US government that cleared the way. When US corporate interests wanted access to the middle east, east asia, south east asia, etc, it was the US government that lent a hand. The US has always been and always will be a mercantilist nation.

> preferring to let market forces have their way.

You are buying into the propaganda. All "market" forces are human forces. After all, there is no market without humans.

> China is tilting the playing field in its favor, while the world's sole superpower is hobbled by clowns and criminals in the executive branch, and corrupt prime contractors pretend to supply the government with tech that they neither understand nor make themselves.

Oh ye of little faith. China has destroyed its environment and sold its people as slave labor to the west for development. If you look at US-china trade, 80% of the wealth is in US hands and 20% of it is in chinese hands.

Take the iPhone as an example. It sells for $800. What percentage of $800 do you think the chinese gets? What percentage do you think US shareholders gets? Almost all of the iPhone's values goes to US based shareholders. The chinese get peanuts.

AKA, industrial policy is not a Chinese invention.


Japan is the master of modern industrial policy. Then South Korea/Taiwan, then China, faithfully adapt the Japanese model.

The iPhone’s value is tied up in its IP (software, design) and high end components, none of which are actually made in China. The assembly happens in China, and the Chinese derive some profit from that, but it is related to the actual value being provided.

>Take the iPhone as an example. It sells for $800. What percentage of $800 do you think the chinese gets?


When the government is your master, you will be innovative only insofar as it is necessary to achieve government objectives. SenseTime may be a great investment because of that implicit support, but it doesn’t make them a great company nor much of a competitive threat to rivals who are exposed to the rigors of the market.

you neglect to mention -- the ability to implement a technology in systems and in the field. Government "by definition" has the ability to use resources to deploy, to a degree that is not easily understood by an individual.

Companies that get too intermingled with the Chinese government are likely to fall victim to some corporate version of Lysenkoism. No company is good at everything. All companies fail eventually. But, if you can’t you’ll eventually grind away miss-applying whatever you can do to whatever you’ve been order to do.

> Certain backward areas have advanced, and various devices always in some way connected with warfare and police espionage have developed, but experiment and invention have largely stopped.

-- George Orwell, "Nineteen-Eightyfour"

That quote is exemplary of Orwell's personal feud with Stalinism rather than an accurate prediction. Historically, neither in the USA or China, has technological cooperation with the military slowed down technological advance (in fact public moonshots were arguably a driver behind most of the technology we take for granted).

For a thorough criticism of 1984 as a device to explore technology in the future, I recommend Asimov's harsh criticism of the book.


The success of Sensetime and Alibaba's (affiliate of Sesame Credit) large investments truly lead me to believe this is a big step towards enforcing the physical aspect of the Social Credit Score, so its not surprising that the government would be all in for helping them along.

My question to your question is, what exactly are we trying to compete towards? I keep reading article after article how China is beating us in AI, but they never explain what that means for us everyday citizens, except for what I can see happening to China. From my perspective, this seems like a frantic push towards omniscient presence in the every day lives of their citizens, which for me completely wipes out any honest benefit AI could provide.

Even China admits that it isn’t beating the USA in AI yet, the articles you are reading, at least if they are from Chinese sources, claim that they will be beating the USA in AI by 2020 or so.

This is sort of a meme with the Chinese government, they make declarations about what will happen in the future, and then silently readjust their goals when the previously declared deadline passes. See for example http://reddit.com/china2020. Think of it as more of a pep rally for a sports match up (rah rah, we will win!).

Forgot the /r/ in the url: https://www.reddit.com/r/china2020

Okay this is off-topic and I apologize, but what's the justification for the self-proclaimed title of the "world's sole superpower?" Is it just a holdover from cold war-era terminology? How is it quantified? How long are we going to make this absurd boast as we inevitably slide towards the fall of Rome, as all great civilizations eventually do?

If it's military might, that may be surprisingly irrelevant as the next century unfolds.

> Is it just a holdover from cold war-era terminology?

This. US and USSR were considered, rightfully, the superpowers during the cold war. "Sole superpower" simply means one has fallen, so the balance has tilt in favor of the US.

It wasn't "self proclaimed". It was globally accepted. It plays into American's patriotism but also flames anti-American in a lot of places. If you think it's just American manufactured claim you're mistaken.

In the same vein, China is considered "new superpower", also rightfully, considering how they have aggressively expanded their influence around the world.

Thanks. Yeah I had a feeling it was an old term from the fall of the Soviet Union. I really still don't understand why the term is still in use, though. It's not the eighties anymore.

>but what's the justification for the self-proclaimed title of the "world's sole superpower?"

The typical justification is that the US is the only nation that unilaterally harasses nations on the other side of the world. When the Russians and Chinese harass other nations those nations are in the same region. When the British feel like making an example of out a country in "our" hemisphere they get permission first.

> Beijing has signaled to investors like Alibaba that it has chosen a winner, and that it will never let SenseTime fail.

Chinese government invested in many startups, the best example is probably Deng Yaping's search engine company around 8 years ago. Such partnership/investment doesn't guarantee anything, Deng's search engine company died in a pretty normal way.

The response from the people I've seen who are quitting Google because of Maven seems to be basically: We should all quit!

Unfortunately this is a collective action problem and if there is even one group who defects, then a boycott is for naught. Might slow things down temporarily but in the end wouldn't stop them.

People who are quitting Google because of maven must be at least a dozen. Out of a company of 80,000. They are a rounding error.

> Not only that, but the US government is neither as aggressive nor as forward-looking about such partnerships, preferring to let market forces have their way

I have not seen this in practice. Wall Street had Glass-Steagall repealed in 1999 which led to the TARP $420 billion bailout in 2008, and Trump just rolled back almost all of Dodd-Frank, so the next taxpayer bank bailout is probably not far off (US taxpayers bailed out S&Ls in the 1980s too). I have not seen much of market forces having their way in the US other than lip service.

Also, look at the 1970s history of Lockheed to see a "never let fail - heed the diktats" company. From the 1971 billion dollar government bailout to the 1976 bribery scandal. Probably one of the most government blessed companies in a country full of such-blessed companies.

This tech enables the Chinese social credit system. Pretty dark use for AI IMHO.

I keep on wondering if the Social Credit system is just for parallel reconstruction of events or not (or maybe both).

Also, people have tweeted at the technology in action.



is a good demo of what they appear to be doing

For the first one they are doing object classification & identification in realtime.

The second one is identifying crowds using a heatmap.

I don't mean to play it down, but honestly classifying objects and identifying crowds don't seem like super valuable outcomes from such a huge investment. I guess I expected something like "persistently label and narrow down the definition of an object over the span of multiple occasions" or "identify the intent or demeanor of a crowd".

They are using this to routinely pull wanted criminals out of crowds using facial recognition. Seems valuable especially to somewhere like China that likes to grab particular people.

Do you realize if you did this at scale how much faster you could look through recordings ? And that this is a manual process currently? Also, they are just starting out .

I just got done reading this New Yorker article yesterday:


China's social credit system is glossed in the article.

Doesn't seem like there are a lot of good outcomes where AI is involved. A passage near the end of the article:

In the meantime, we need a Plan B. Bostrom’s [author of book Superintelligence] starts with an effort to slow the race to create an A.G.I. [Artificial General Intelligence] in order to allow more time for precautionary trouble-shooting. Astoundingly, however, he advises that, once the A.G.I. arrives, we give it the utmost possible deference. Not only should we listen to the machine; we should ask it to figure out what we want. The misalignment-of-goals problem would seem to make that extremely risky, but Bostrom believes that trying to negotiate the terms of our surrender is better than the alternative, which is relying on ourselves, “foolish, ignorant, and narrow-minded that we are.” Tegmark [author of book Life 3.0: Being Human in the Age of Artificial Intelligence] also concludes that we should inch toward an A.G.I. It’s the only way to extend meaning in the universe that gave life to us: “Without technology, our human extinction is imminent in the cosmic context of tens of billions of years, rendering the entire drama of life in our Universe merely a brief and transient flash of beauty.” We are the analog prelude to the digital main event.

Takes the idea of moving fast and breaking things to the next level.

Wow, for anyone who hasn't read into this, check out the Wikipedia page[1]. Feels like sci-fi dystopian futures are closer everyday.

[1] https://en.wikipedia.org/wiki/Social_Credit_System

I think we are already in a Dystonian future and people are distracted and don't notice.

Vast prison population and byzantine court system, Declining life expectancy, Horrific traffic, Surveillance state including 24/7 location tracking, Births below replacement rate, Systemic Unemployment and Age Discrimination, etc

Individually they don't seem bad until one of them bites you. But, the world is a very nasty place for many people.

Oh yeah, this is dystopia baby. For many, at least. Like everything, it's probably on a spectrum and it's probably degenerative over time.

Lots of people will point to how prosperous we are and how things like premature death and disease are at record-low rates throughout all of history.

I think we're in a nice little bubble of time where we're surfing on a false economy, borrowing from our future selves in a very unsustainable way which artificially makes our current situation look just about fine. But eventually we're going to use up this runway. At risk of mixing metaphors haphazardly, this is sort of the calm before the storm.

As climate change starts ramping up this shit is going to get very uncomfortably real, and lots of these nightmare dystopia-technologies we're currently developing will probably start seeing even greater mass usage.

I'm typically not the apocalyptic type, and I recognize that most members of every generation that's ever lived have believed that they were one of the last generations. But our universe has shown us that eventually everything ends, so realistically there will someday be a generation which makes this assumption correctly. I don't think we're necessarily they, but we might be their grandparents or great-grandparents.

There's a non-trivial chance that we've already passed the point of no return on climate change. We've almost casually walked into the maw of the great filter already--it just hasn't chomped down on us yet.

- Vast prison population - Declining life expectancy - Horrific traffic

These 3 at least are a very american-centric view of things. In our good old Europe (excluding UK) things are way better.

Now, some things on your list are universal :/…

This is completely tangential, but adds to the sort of vision of the world on that stage. Specifically your reference to how hard the world can be for some people.

It's a brilliant and beautifully photographed documentary about the people who ride the cross-Saharan train, sometimes as the core of their livelihood, and the people who rely on this older technology just to be able to eat.

It's titled The Mauritania Railway: Backbone of the Sahara


edit: I just can't recommend it enough. Just watching a few moments already has me sucked in

To me, the whole present situation seems much closer to the story line in Elysium. I've worked in factories like the protagonist. It's not as distant as it can seem.

> Births below replacement rate

This is the opposite of a problem. Overpopulation is a huge global issue.

> Vast prison population and byzantine court system > Systemic Unemployment and Age Discrimination

You forgot discrimination based on ethnicity, nationality and, most notably, wealth.

Birth below replacement rate being a problem is usually in the context of the problem of "us dwindling versus them proliferating".

Every large ethnic group is at least slightly below replacement rate in the USA. Some more than others but it's not a good sign suggesting that people are generally struggling.

PS: Male:Female gender imbalance at birth + early deaths is why the replacement rate is over 2.0 lifetime births per woman.

The future is already here — it’s just not evenly distributed’. William Gibson

Not sure "Births below replacement rate" is dystopian, "uncontrolled and unsustainable population growth" definitely is. Sadly, what we have is the latter

> "uncontrolled and unsustainable population growth" definitely is. Sadly, what we have is the latter

No we don't. IIRC, birth rates are dropping everywhere and the world population in on track to plateau mid-century before it declines.

And also "uncontrolled and unsustainable" living standards growth. When you have huge portions of world population wanting to live like in the west, and not only for basic stuff like hygiene and schooling. And meanwhile people are decrying that we're too many, too many people are born. Which basically translates to more women should take more pills. Children of men, the movie, comes to mind.

Wow, given this and the AI system it sounds way too much like Psycho-Pass. Pretty horrific indeed.

I read the Wikipedia entry and I immediately thought of Sibyl. It's horrifying and quite sad knowing that many people predicted this long ago and we did nothing but rejoice for the literary excellence of their works (e.g. 1984)

Maybe it's the communist ideals, or maybe it's the reality of governing over one hundred cities with populations over one million, but personal liberty is valued less than social order and economic advancement in the Chinese goverment.

I think it's more dystopian for all governments to have the same ideals and policies than for different governments to have different ideals and policies. It's okay that China is different.

Yes. This Chinese social credit score undermines the free market in social credit scores, leading to inefficiency and lost opportunities.

Given that most of the value of the system is derived from centralisation of data and applying the exact same standard to every citizen to ensure comparability the free market in this case would not increase efficiency, in the same way healthcare markets don't.

Multiple actors would just increase bureaucratic overhead.

Right? I suppose any stupendously powerful technology that isn't yet adequately understood or controlled will tend towards dark uses.

AI might be significantly composed of hype today, and there may indeed be an AI winter on the horizon as people start to trade in insane optimism for realistic expectations, but if a little of the hype is true then we have likely completed the equivalent of a Manhattan project to develop something truly nuclear in its potential to disrupt human civilization.

In the twentieth century we found a way to leverage physical energies greater than we'd ever imagined humans could wield. Now we might be developing something more subtle, yet also insanely powerful: a way to leverage data manipulation in a way that will slowly transpose democracy and individual thought through brute analytical power.

Or so the doomsaying goes. It may be baseless paranoia, but it's credible enough to deserve some very strong caution. From what we've seen in our universe we're probably due to run into a great filter very damn soon. It's probably something we already know about today, in some form at least.

Mass surveillance is quite possibly the most practical large-scale application of modern AI.

"Some people just don't understand the dangers of indiscriminate surveillance." -JC Denton, Deus Ex

China is on track to have the most advanced systems of mass surveillance in the world.

I've seen quite a lot of criticism against the social credit system. But to me it it just crime history and (financial) credit history mined from big data and combined into a centralized system.

What's particularly bad about the system per se, without getting into the arguments specific to Chinese politics?

> I've seen quite a lot of criticism against the social credit system. But to me it it just crime history and (financial) credit history mined from big data and combined into a centralized system.

> What's particularly bad about the system per se,

Tightening everyday social control that's at least partially designed to serve the political goals of an unabashedly authoritarian regime that's self-consciously opposed to things like "universal values of human rights." [1] [2]

It might be easier to understand it this way: in China, you're a de-facto criminal if they judge you to be in some way opposed to the regime.

[1] https://www.nytimes.com/2013/08/20/world/asia/chinas-new-lea...

[2] https://www.nytimes.com/2017/07/25/magazine/the-lonely-crusa...

> without getting into the arguments specific to Chinese politics?

That's almost like asking what's bad about something without getting into its bad aspects.

I haven't read much about it, but isn't China's social credit system (sesame credit) based off of Aliexpress and NOT from the government like a lot of people are presuming? When people say Chinese social credit system, it sounds like it's the government that owns and enforces it.


Could you please stop posting unsubstantive comments to Hacker News? Above all on flamebait topics.

Unsubstantive comments on flamebait topics = de facto trolling.

How can't this be a legitimate opinion?

That's beside the point. The point is that the comment contains no more information than "yay flamebait". The signal/noise ratio of HN comments needs to be much higher.

Unsubstantive comment + flamebait topic = de facto trolling. By de facto I mean whether you intended to troll or not. We can't read your mind, but we know a lot about the likely effects of such a post. Moreover, you have a history of doing this, e.g. https://news.ycombinator.com/item?id=17102050. Please don't do it any more.

So you are saying HN has taken an active stance against Islam?

If you continue to break the site guidelines we will ban you.

Who needs social credit scores telling you what to think when you have HN!

Can you provide any details as to what you define as good, and how this system reflects that?

Off topic, but just looking through TechCrunchs GDPR options, they list at least 50 'advertising partners' who they work with to personalise your ads.

I don't work for a company who runs sites like these, but is that normal? How do you maintain a relationship with 50 different ad vendors, or do they actually just deal with 2-3 and then there's a whole graph of others on some sort of affiliate scheme?

I imagine that's what you would get with a competitive marketing team, where everyone would handle one or more of those relationships.

The latter. It’s like the recursive junk you collect by installing a single npm package.

Powered by Caffe, an open-source framework developed at Berkeley. The irony.

Developed at Berkeley by Yangqing Jia, a Chinese national.

A Chinese national whose project was backed with US public funds. Triple irony.

There are (many) more Chinese nobel prize winners who won them for work in the USA than in China. It doesn’t sound like irony.

A large portion of US inventions were made by people of different nationals, I don't see the irony here but only narrow-mindedness.

I'm not objecting to foreign nationals doing publicly funded research in the US. And I was not the first to bring up his nationality. Personally, I don't care what his nationality is. The initial irony I noted was that a US university had produced an open source project that is the chief tool of face detection on the mainland.

I don't get the irony. Does nationality matter for open source project with a clear license?

double irony rainbow!

With this much money, surely things will end well and this money with be used prudently to advance their future.

You forgot the “/s”

I don't think we can say the AI Winter is here until these large funding rounds stop.

I think that even if research just stopped (like in an AI winter), there's still a lot of applications for the tech we have now. I suspect that's where most of this money will go.

I don’t think there will be an AI winter frankly. A lot of the use cases that businesses want can be built with a combination of decision trees and some ML for classification.

ML works incredibly well on even lightly-structured data (which includes most financial data streams). Agree it hasn’t lived up to the AI hype, but that doesn’t mean what it can deliver on isn’t useful.

I don’t think we can say that the AI winter is coming without these large funding rounds.

From the author of the AI Winter post:

"I don't think the winter depends on the amount that somebody makes today on AI, rather on how much people are expecting to make in the future. If these don't match, there will be a winter. My take is that there is a huge bet against the future. And if DL ends up bringing just as much profit as it does today, interest will die very, very quickly."

This possibility is interesting to consider from an economic perspective.

Let's posit that DL does just that, bring in the same value (as a whole system) as non-DL.

Inside such a system, there is a business advantage to implementation of DL, and there is a concentration of value around those with the largest dataset.

This means that DL systems will still be very valuable in that they are a competitive advantage, but only so much as they are difficult to implement independently.

The value differentiator then becomes the data itself. I think this points to a value proposition for data providers. GDPR and such seem to be quite correctly identifying and correcting for this risk.

These funding rounds just mean we are using existing tools. The AI winter is regards to discovering new AI techniques, not re-applying existing discoveries to business.

> These funding rounds just mean we are using existing tools. The AI winter is regards to discovering new AI techniques, not re-applying existing discoveries to business.

There's so much hype around "AI" right now that I wonder if we're headed to a winter where you need very strong business justification to get funding to apply those existing tools. It seems like anyone with a problem to solve is spending millions to throw machine learning at it, just to see if it will stick. That's not sustainable.

AI Winter was referring to how the potential of current AI methods is limited, and aren't going to naturally evolve into artificial general intelligence. However I don't think it matters in this case. There are still many untapped applications of the AI technologies that work today. Deploying recognition technologies at scale in China to strengthen the government's control over the populace is one of them.

> AI Winter was referring to how the potential of current AI methods is limited

Not exactly. "AI Winter" refers the historical periods that followed unsustainable hype that surrounds AI methods prior to those times. Skepticism of AI during the last winter was so strong that "AI" kinda became a dirty word and people invented euphemisms like "machine learning" to avoid the stigma.


More data (the copious amounts of data China is gathering on its people) doesn't itself make AI technology better - or does it, theoretically? Or put it this way: I'm assuming the real revolution in AI will be from talented-ass scientists and algorithms, not just from the amount of data you throw at computers.

A lot of the algorithms center around enabling the training of more data (e.g. one bit DAC for distributes GPU training).

For a lot of problems, more data is the only solution.

My comment on SenseTime

It is a unicorn of few billion bucks without a SINGLE THING that makes money.

They have an "AI" supposedly, but that thing does not make money magically appear in their pocket.

At best, they market a face recognition library, that sucks in comparison to a first grader's OpenCV hello world.

AI might be one of the few tech niches where China has a legitimate advantage. Obtaining certain kinds of data is tough, there are many unresolved ethical and legal considerations. The communist regime has already proven itself to have no moral qualms about the most abhorrent acts like organ harvesting from political prisoners. Rule of law is not a problem either. The regime has also displayed great interest and belief in AI. Whatever the companies want they're gonna get.

skynet in action

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact