The world’s most unproductive entrepreneur

The new book by Wall Street Journal reporters Tom Wright & Bradley Hope, Billion Dollar Whale: The Man Who Fooled Wall Street, Hollywood, and the Worldis amazing. It must be one of the most detailed accounts of high-level corruption ever produced. I had followed their reporting on Malaysia’s 1MDB scandal, and generally understood that the financier Jho Low had colluded with former prime minister Najib Razak to embezzle ungodly amounts of money from the state development fund.

But what I had not really grasped before reading this book is that Low did not just steal money that belonged to the Malaysian state and people–he actually organized the creation of the 1MDB fund, and effectively controlled it. The 1MDB fund in fact never had any other purpose or rationale than as a vehicle for corruption.  The scheme got its start when Low approached the sultan of Terengganu, cleverly exploiting the opportunities for corruption in Malaysia’s system of hereditary monarchies:

He’d built a quick reputation in Malaysia as a deal maker, but, as always, Low was keenly aware of how his success stacked up on a global stage. Low had observed the power and status of Khaldoon Al Mubarak of Mubadala, who ran the Emirati sovereign wealth fund. A fund like that had billions of dollars in investments, not mere millions.

Why, Jho Low wondered, couldn’t he put together a sovereign wealth fund of his own—one based in Malaysia? But where could he find the initial funds? Traditional sovereign wealth funds invest oil profits, and so Low honed in on the Malaysian state of Terengganu, which was rich in offshore oil and gas fields.

But it didn’t take long for that local fund to become a national initiative. In August 2009, Najib announced the launch of the 1MDB fund at a dinner with the crown prince of Abu Dhabi:

The 1MDB fund was simply the Terengganu Investment Authority, which had recently raised $1.4 billion in Islamic bonds, transformed into a federal entity. The 1MDB fund would be responsible for repaying the bonds. Once Najib came to power, Low convinced him to take over the fund, broaden its remit, and look for Middle East backing.

Low’s pitch for the 1MDB fund was basically that it would be a completely unaccountable vehicle for the prime minister to dish out development projects and political patronage to benefit himself and his supporters:

The 1MDB fund was supposed to invest in green energy and tourism to create high-quality jobs for all Malaysians, whether of Malay, Indian, or Chinese heritage, hence the slogan “1Malaysia.” The fund, Low promised the prime minister, would suck in money from the Middle East and borrow more from global markets. But he had another selling point, one which Najib, who was ambitious, found extremely attractive: Why not also use the fund as a political-financing vehicle? Profits from 1MDB would fill a war chest that Najib could use to pay off political supporters and voters, restoring UMNO’s popularity, Low promised.

Over the next few years, Low and his conspirators would use 1MDB to borrow billions of dollars from global capital markets, and would simply take much of the proceeds for themselves. To achieve all this required Low to do constant networking, set up immensely complicated financial and legal arrangements, and splash out tens of millions of dollars along the way to win friends and influence people. Low in fact worked hard at his corruption, and displayed unmistakable entrepreneurial flair–a huge amount of effort deployed to make his country poorer not richer.

Previously I had thought of corruption as mostly acting like a drag or tax on the economy: the extraction of bribes to approve a construction permit or dismiss traffic fines. My imagination had not quite encompassed the idea that corruption could be a motivating force for huge economic initiatives that would be entirely wasteful. So one lesson from the book seems to be that I have been insufficiently cynical about human frailty.

Low’s tale also made me recall the distinction that William Baumol made between productive and unproductive entrepreneurs, in a famous 1990 article:

The basic hypothesis is that, while the total supply of entrepreneurs varies among societies, the productive contribution of the society’s entrepreneurial activities varies much more because of their allocation between productive activities such as innovation and largely unproductive activities such as rent seeking or organized crime. This allocation is heavily influenced by the relative payoffs society offers to such activities. …

If entrepreneurs are defined, simply, to be persons who are ingenious and creative in finding ways that add to their own wealth, power, and prestige, then it is to be expected that not all of them will be overly concerned with whether an activity that achieves these goals adds much or little to the social product or, for that matter, even whether it is an actual impediment to production (this notion goes back, at least, to Veblen). …

There are a variety of roles among which the entrepreneur’s efforts can be reallocated, and some of those roles do not follow the constructive and innovative script that is conventionally attributed to that person. Indeed, at times the entrepreneur may even lead a parasitical existence that is actually damaging to the economy.

Baumol’s article cited many interesting historical examples of unproductive entrepreneurship, but his contemporary examples were all a bit dull: he mentions tax evasion and corporate lawsuits. Low’s audacious scheme to create a sovereign wealth fund in order to plunder it surely makes him the paragon of twenty-first-century unproductive entrepreneurship.

51b7qwohbxl-_sx329_bo1204203200_

What I’ve been listening to lately

  • Hank Mobley – And His All Stars. The perenially underrated tenor saxophonist Hank Mobley recorded a string of interchangeably-titled dates in the mid-50s, all of which have their moments. But this 1957 session is really a cut above thanks to the presence of Milt Jackson, who lifts the proceedings to another level. Just a sterling example of mainstream jazz.

  • Julian Lage – Live In Los Angeles. Lage is a virtuosic young jazz guitarist whose twangy sound pays homage to the country and rock traditions. He won my heart by tackling early jazz obscurities like “Persian Rug,” which I know from a 1928 recording with Fats Waller on organ. But he is not at all a moldy fig–these are modern, free-flowing improvisations. The extended live versions to me are better and more energetic than the ones on his studio album; the whole thing is free on Youtube.
  • Bessie Smith – Empress of the Blues: Volume 2, 1926-33. It turns out I have been a Bessie Smith fan for years without even knowing it. Many of my favorite classic blues tunes that I know from other singers–“Gimme A Pigfoot And A Bottle of Beer”, “Careless Love”, “I Need A Little Sugar In My Bowl”, “Nobody Knows You When You’re Down and Out”–were made famous by her, but I had not sought out the original recordings until now. The sound on this JSP collection is quite impressive, Smith’s incredibly strong voice just punches through.
  • Harry Miller – 1941-1983: The Collection. It’s a bit unfair to put this on a list of recommendations since it is long out of print, but I felt compelled to share since it is by far the best music find I have ever made at a thrift store. Miller was a South African bassist who lived in London, and his work straddles British free jazz and the more groove-oriented South African scene. He has strong groups on Family Affair and Down South which are the most consistent albums, but his solo bass recording Children At Play is also interesting. A live recording from this period is available from the essential Cuneiform Records.
  • The Sun Ra Arkestra – Live At Babylon. The music produced by the current edition of the band, under the leadership of Marshall Allen, is perhaps not as complex and weird as it was under the master himself, but that is really just a quibble. This 2016 live recording captures the Arkestra’s ferocious groove and exuberant melding of the tonal and atonal; the long version of “Discipline 27” is just a monster.

Ideology and the sources of policy error

Stephen Joske has an essay at War on the Rocks that makes a good point about the possibility of a Chinese financial crisis:

We commonly hear that China cannot have a financial crisis because the government owns all the banks and can control them. … In theory, government control is good for stability. In practice, however, two things have to happen to avoid a crisis: First, the government has to use its power to make the right policy choice, and second, it has to avoid making a Lehman-style regulatory mistake. …

While the government owns the banks, that does not stop officials from making regulatory mistakes. We have already seen regulatory mistakes such as mishandling of RMB market volatility in 2015. We have also seen well-handled, timely, and complex crisis management failing financial companies. But that does not mean they always handle every crisis well and will continue to do so. The law of probability indicates that eventually something will go wrong and, like every other country, China will have a financial crisis.

It is a fair point that humans are fallible and will eventually make a mistake, and we should not presume that the Chinese financial authorities are more or less fallible than all other humans.

But I think it is a bit of a cop-out to just say, oh, policy errors happen, and not to inquire more deeply into the actual causes of policy error. I think if we look at the recent history of financial policy mistakes, many them were in fact not driven by random errors but by ideology.

It is hard to separate the US policy errors in the global financial crisis from the strong ideological value placed on the government abstaining from intervening in markets. The day after Lehman Brothers failed, then-St. Louis Fed president Andrew Bullard said: “By denying funding to Lehman suitors, the Fed has begun to reestablish the idea that markets should not expect help at each difficult juncture.”

There would have been technical and legal difficulties to mounting a rescue of Lehman, but these could certainly have been surmounted if officials had been really worried about the impact of a failure on the financial system. They were not that worried (which in hindsight was clearly an error) and were preoccupied with sending signals about the correct relationship between the government and the financial system.

Most financial crises take the form of a “run on the bank,” even if the bank is not called a bank and those doing the running are not retail depositors. When investors no longer wish to hold bank liabilities, the bank cannot fund itself and experiences a liquidity crisis. But if investors know or believe that the government will rescue the financial institution, they are less likely to be so fearful of failure as to stop holding its liabilities, and less likely to fear that the problems of one institution will spread to others.

So a system in which the government is ideologically committed to maintaining a separation between the public sector and a private financial sector, and a system in which the government is ideologically committed to maintaining the union of the public sector and the financial sector, are not at all equivalent in terms of financial risk. In the former, there is much more of a risk that private investors will become worried about the failure of banks and stop funding them. (The latter, of course, has the problem that banks will tend to abuse government support and take more risk than they should.)

So it seems to me that, while Chinese officials are not more or less fallible than other humans, they do operate in a system that tends to minimize one major source of the policy errors that tend to lead to financial crises. Simply put, Chinese officials are more likely to err in favor of providing government support to a failing institution than in favor of denying it. This does not mean that Chinese banks cannot fail–it seems quite likely to me that some smaller institutions will eventually go under–but that such failures are much less likely to cause contagion and a crisis of confidence in the financial system.

Arguably Chinese financial policy is in fact less ideological than American. That may seem a strange thing to say about a system in which ideology is highly formalized and there is a large institutional structure for ensuring compliance with ideology. But the ideology of the Communist Party is inflexible only about ends, rather than means. Continuation of Party rule and its ability to direct social and economic development cannot be questioned.

But China is fairly flexible about the means employed to achieve those ends. American political ideology by contrast puts strong constraints on means: there are often intense arguments about why various things cannot be done or should be done in order to maintain the self-image of the US as a free-market economy.

The US-China trade war as a conflict of values

The tariffs that the Trump administration has imposed on Chinese goods are seen by the Chinese government as unprovoked and unjustified assaults. So there are few opinions more unwelcome right now than that China brought the trade war on itself. Yet that is more or less what a couple of Chinese liberal intellectual are saying openly: that trade conflict with the West is the inevitable result of China’s promotion of a state-centered development model.

One of these voices is Sheng Hong of the well-known Unirule Institute, who published an interesting article on US-China relations on FT Chinese on October 19. The Unirule website has helpfully provided an English translation, but I have re-translated the portions below myself for greater clarity:

China’s reform and opening up is the guarantee of strategic cooperation between China and the US. Such strategic cooperative relations would never have been possible if China was still stuck in the Cultural Revolution, when it practiced class struggle and a planned economy domestically, and exported revolution abroad. Reform and opening up not only brought people economic freedom, but also changes in the political structure. The emancipation of thought has to some extent loosened controls over the freedom of speech, and the freedom of the market economy has allowed people to throw off the shackles of their work units. As the market played a greater role in more areas it reduced government’s direct control over society. Implementing market regulation relied on a just legal system. Only a China that is in this way progressing toward marketization, rule of law and democratization can be accepted by the US on a strategic level, and create the framework for strategic cooperation with the US.

Without a doubt, reform and opening up eliminated the ideological conflict between China and the US, as well as the whole Western world, and gradually brought convergence in terms of values. … Some of the so-called “socialist core values” promoted by the Chinese Communist Party overlap with values recognized by the US and the western world, for instance freedom, democracy, rule of law, equality and justice; while other values, such as civilization, harmony, integrity and dedication, are not in conflict with the values of the US and the western world.

We must clearly recognize that such convergence of values is the basis for strategic cooperation between the US and China. Only through a convergence of values can China be deemed a factor of peace and stability in international relations, and be seen as a trustworthy nation with which close cooperation is possible, rather than one that proclaims the overthrow of the capitalist world and does not renounce the use of violence. Only such a country that advocates peaceful means to resolve disputes between nations,  and does not resort to the threat or use of force, can provide the world with a stable and just international order.

Therefore, China’s reform and opening-up and the China-US strategic cooperation are two inter-related things. That is to say, there is no strategic cooperation without reform and opening-up, nor is there reform and opening up without China-US strategic cooperation. … China should not, and cannot, seek hegemony over the world by going against the rules of civilization accepted by China and most of the countries in the world. Only on the basis of respecting the consensus rules of human civilization can China overcome the mistakes and deviations of the US, and become a civilizational center with moral legitimacy and great economic strength. Today, China faces the risk of leaving the path of reform and opening up, which would risk the loss of strategic cooperative relations with the US. Such a result would be a complete failure.

ShengHong

Sheng Hong

Somewhat similar sentiments appear in an October 14 speech by Zhang Weiying, a prominent liberal economist at Peking University (liberal in the Chinese sense of favoring free-market policies; Zhang is more of a Hayekian libertarian). The following is my translation of some portions of a summary that was posted on the website of the National School of Development (it was later removed after attracting press coverage).

The rapid development of China’s economy and the improvement of people’s living standards over the past 40 years are facts denied by no one, but there is still controversy over how to understand and interpret these facts. At present, there are two interpretations of China’s growth in the past few decades, the theory of the “Chinese model” and the theory of the “universal model.” The former holds that China’s economic development benefited from a unique Chinese model, with a strong government, many large state-owned enterprises, and wise industrial policy.

The latter holds that China’s remarkable achievements are, just as with the rise of Britain, France, postwar Germany, Japan and the Asian tigers, based on the power of the market and the creative and risk-taking entrepreneurial spirit. China also made use of the technologies accumulated by Western developed countries over the past three hundred years. As I explained in an article published early this year, China has in the past 40 years of reform and opening up experienced the three industrial revolutions that took the Western world 250 years. The latecomer’s advantage means that we have avoided many detours and directly share the technological achievements that others have already obtained through experiments of great cost. …

The above evidence shows that the theory of the “Chinese model” is seriously inconsistent with the facts. China’s high growth over the past 40 years has come from marketization, entrepreneurship and the technology accumulated by the West over three hundred years. The bigger problem is that using the “Chinese model” to explain the achievements of the past 40 years is not beneficial for China’s future development.

Domestically, misleading yourself means a future of self-destruction. Blindly emphasizing the unique Chinese model means going down the road of strengthening state-owned enterprises, expanding government power, and relying on industrial policy. This will lead to a reversal of the reform process, the abandonment of our predecessors’ great cause of reform, and ultimately economic stagnation.

Externally, misleading the world leads to confrontation. From the Western perspective, the “China model” theory makes China into an alarming outlier, and must lead to conflict between China and the Western world. The unfriendly international environment we face today is not unrelated to the mistaken interpretation of China’s achievements over the past 40 years by some economists (both Chinese and foreign). In the eyes of Westerners, the so-called “China model” is “state capitalism,” which is incompatible with fair trade and world peace and must not be allowed to be advance triumphantly without impediment.

291205348hwf

Zhang Weiying

In some ways it is not surprising to hear such statements from Sheng and Zhang, whose views are well established, and also far out of the mainstream of Chinese intellectual opinion. What is interesting is that these views are coming out at this moment–although since the comments of both authors are regularly scrubbed from the Chinese internet, it is hard to know how much impact they have.

Gessen on Shalamov: “It was nothing personal. Just the twentieth century.”

I really enjoyed Keith Gessen’s new novel A Terrible Countrya charming, convincing and moving account of a young American’s complicated relationship with Russia. The narrator is a struggling low-level academic specializing in Slavic literature, who must return to Moscow to care for his aging grandmother. He gets by teaching an online course in Russian literature, but generally seems more passionate about playing pick-up hockey than the great Russian books.

One of the only exceptions, to my surprise and delight, is a digression on Varlam Shalamov’s Kolyma Stories, which I just read this year and immediately recognized as an indisputable masterpiece. Here is the narrator’s appreciation of Shalamov:

Shalamov saw things differently from Solzhenitsyn. He saw them doubly, ambivalently. He thought Solzhenitsyn was a windbag. Physical pain, hunger, and bitter cold: these could not be “overcome” by the spirit. Nor did the world divide neatly, as it did for Solzhenitsyn, between friends and enemies of the regime.

For Shalamov, in the camps there were people who helped him and there were people who brought him harm (who beat him, stole his food, ratted him out), but the majority of the people he encountered did neither. They were just, like him, trying to survive. There was great brutality in the camps, and very little heroism.

In his memoirs he told a remarkable story about learning, at one of the darkest moments of his camp life, that his sister-in-law, Asya, with whom he was close, was in a nearby camp. Shalamov was in the hospital with dysentery, and one of the doctors wanted to know if he wished to send Asya a message. Only half alive, Shalamov scribbled her a short and unsentimental note. “Asya,” it said, “I’m very sick. Send some tobacco.” That was all.

Shalamov clearly remembered this with shame, but also with understanding: he was weak, on the edge of death, and had been reduced to a bare animal existence. There was no great lesson in this, except that in certain conditions a man quickly ceases to be a man.

It was nothing personal, as the saying goes. Just the twentieth century.

This is very well put, far better than I managed when trying to express what is so great about Kolyma Stories (here is my previous post on Shalamov).

For more on Gessen’s novel, Francise Prose’s review is on the mark, and gives a good flavor of what the book is like.

a1nynnrmhhl

What I’ve been listening to lately

  • Johnny Dodds – Blue Clarinet Stomp. Dodds was the clarinetist in Louis Armstrong’s classic Hot Fives sessions, and these 1928-29 recordings on his own capture some of that same powerful early jazz magic. Particularly wonderful are the trios with just piano and bass; bassist Bill Johnson plays out front and melodically in a way that would not become the norm for jazz bassists for another generation.
  • Vijay Iyer & Wadada Leo Smith – A Cosmic Rhythm With Each Stroke. An absolutely entrancing duet. Iyer is a great partner for Smith’s gorgeous long tones and masterful use of space. I find Smith’s music hard to describe but increasingly rewarding.
  • Don Ellis – Essence. Ellis is best known for his 1970s big-band work which featured odd time signatures, but this stuff is, justifiably, “receding from historical view,” as Ethan Iverson put it. More enduring are Ellis’ first recordings in the early 1960s with some adventurous small groups. So far this one with Paul Bley and Gary Peacock is my favorite of the lot: Ellis really rips into some well-chosen standards and modern compositions.
  • Jakob Bro – BalladeeringI first heard this one in, of all places, that London bookshop on a boat by King’s Cross. It stood out then, as it should have, given the caliber of the bandmates the Danish guitarist found: Lee Konitz, Paul Motian, Bill Frisell. All of them excel at creating a spacious, almost ambient sound.
  • Jimmie Rowles – The Peacocks. A really lovely recording with Stan Getz; there are quartet and vocal numbers, but the real standouts are the duets, including the title track. Discovered via the amazing Ted Gioia, who is a Rowles advocate.

Seizing the moment for artificial intelligence: my take on the US-China rivalry

Today I am republishing a piece originally written for Gavekal Dragonomics clients a few months ago, on the US-China rivalry in artificial intelligence. Usually these pieces stay behind the paywall, but with our partner company Evergreen Gavekal we’re making it available for a general audience.

I wrote this piece as a way of sorting out of my own thinking on how to place recent technological trends in the broader story of China’s economic development. I make no claim to be an expert on artificial intelligence: these are just the thoughts of a China watcher trying to absorb what the technologists are saying.

The full text follows below, or you can download the PDF:

 

Seizing The Moment For Artificial Intelligence

by Andrew Batson (originally published May 3, 2018)

In the escalating trade dispute between the US and China, technology has increasingly become the key issue, overwhelming more traditional economic topics like tariffs, deficits and currency valuations. Both countries see their economic future as depending on their success in high technology, and each is worried they will lose out to the other. One of the most intense areas of focus is artificial intelligence, where recent rapid breakthroughs have captivated investors and the media—and where China has emerged as the main US rival.

In this piece, I will try to provide some nontechnical answers to the questions of the moment: What is artificial intelligence anyway, and why is it a hot topic? Why does China seem to be doing so well in artificial intelligence? And how should we think about the rivalry between the US and China to develop this technology? In brief, I think China will do well in artificial intelligence, in part because the technology is now in a phase that plays to its strengths. But it does not make sense to think of the US and China being engaged in an artificial-intelligence “race” along the lines of the US-Soviet space race.

Machines don’t think, but can do useful stuff

A common-sense definition of artificial intelligence, what a layman might understand the term to mean on the basis of reading the news and watching television, is probably something along the lines of: machines that can think like human beings. Artificial intelligence in this sense does not exist, and according to most researchers in the field there is no prospect of it coming into existence any time soon. There are enormous philosophical and technical challenges in understanding how human minds work and replicating those processes in software, and most of these challenges have not been solved.

A more precise definition of artificial intelligence, closer to that used in the industry, would be: the development of computer systems that can perform tasks associated with human intelligence, such as understanding speech, playing games, or driving cars. In a way the term is misleading, because what is being mimicked is not human intelligence itself, but the practical results of intelligence being applied in specific contexts. Artificial intelligence in this narrow sense does very much exist, and its progress is now attracting feverish interest from business, venture capitalists, governments and the media.

The turning point seems to have come around 2014-15. Since then, software programs have been able to match or exceed human performance in tasks that previously could not be reliably performed by machines: recognizing faces, transcribing spoken words, playing complex games. One landmark that had particular resonance in China was the 2016 victory of Google’s AlphaGo software over a master South Korean player of the board game Go (known as weiqi in Chinese); AlphaGo subsequently also defeated China’s top-ranked player. While chess programs have been beating human masters for years, Go is much more complex; the number of potential board positions is traditionally estimated at 10172, more than the number of atoms in the universe.

How are these feats possible? Most of what is referred to as “artificial intelligence” in the media is a subset of the field known as machine learning, and in particular a subset of machine learning called deep learning. All software works by following clearly specified instructions, known as algorithms, on how to perform a specific task. In machine learning, the algorithms are not fixed in advance, but evolve over time by building data-driven models.

Usually the type of software used in machine learning is called a neural network, because its structure is loosely inspired by the connections between neurons in a human brain. The network takes an input signal and repeatedly processes it into something more useful; what makes the learning “deep” is that there are a large number of “layers” that process the signal. The use of these techniques means that an algorithm can improve its performance of a task by repeated exposure to data. They are particularly useful where writing algorithms the traditional way—by specifying all possible details and eventualities in advance—is cumbersome or impossible. The concept of machine learning dates to the 1960s, and much of the original work underlying today’s approach of deep learning dates to the 1980s and 1990s.

More power, more data

The rapid improvement in the results of specific machine-learning applications in recent years is thus not a result of fresh theoretical breakthroughs. Rather, it has happened because advances in computing power have allowed machine-learning algorithms to run much faster, and the increased availability of very large amounts of structured data have given them much more to work with. The lesson has been that lots of processing of lots of data is required for the algorithms to be effective in finding patterns. This in itself is a sign that machine learning is not very much like human learning: humans can learn quickly from small numbers of examples, by building internal mental models. Machine learning by contrast is a massive and repetitive number-crunching exercise of building up statistical regularities.

Researchers in the field sometimes describe machine-learning algorithms as being “narrow” and “brittle.” Narrow means that an algorithm trained to solve one problem in one dataset does not develop general competencies that allow it to solve another problem in another dataset; an algorithm has to be trained separately for each problem. The Go-playing algorithm is not also capable of analyzing MRI scans. Brittle means that the algorithm only knows its dataset, and can break down if confronted with real-world situations not well represented in the data it learned from. An often-used example is facial-recognition software that is trained on databases consisting largely of photos of white men, which then fails to accurately recognize faces of black women.

But while it is important to understand that machine learning is not a magic wand, it would also not do to underestimate its potential. Machine learning is essentially a way of building better algorithms. That means it could be applied to almost any process that already uses software—which, in today’s world, is quite a lot—as well as many new processes that could not be effectively automated before. The most obvious example is self-driving cars, which can already operate in restricted contexts and could be in general use within a decade. Machine learning is already being used to spot patterns that previously required trained human expertise, such as recognizing financial fraud or early-stage cancers. Because of this broad applicability, enthusiasts call machine learning a “general purpose technology” that, like electricity a century ago, can boost productivity across every part of the economy.

Throwing resources at the problem

The key point is that machine learning has now moved from a pure research phase into a practical development phase. According to Oren Etzioni of the Allen Institute for Artificial Intelligence, all of the major recent successes in machine learning have followed the same template: apply machine-learning algorithms to a large set of carefully categorized data to solve problems in which there is a clear definition of success and failure.

All parts of this procedure are quite resource-intensive. Huge amounts of computing power are required to run the algorithms. The algorithms need huge quantities of data to find patterns. That data must also be first carefully structured and labeled so that the algorithms can draw the right conclusions—for instance, labeling pictures of objects to train an image-recognition algorithm—a process that is extremely labor-intensive. The repeated training and refining of the algorithms also requires a lot of labor by highly skilled workers, whose numbers are necessarily limited. But the reason for the excitement over artificial intelligence is that there is a now a sense that the main remaining constraints on progress are these limitations of resources—and such limitations will be solved over time.

According to its many boosters, China has all of the necessary resources to make progress in machine learning. It has large and well-funded technology companies, including publicly traded giants like Tencent, Alibaba and Baidu, but also private companies with multi-billion-dollar valuations like ByteDance, which runs a popular news app with personalized recommendations, and SenseTime, which specializes in image and facial recognition.

AI-chart1

China also has the world’s largest population of internet and mobile phone users, who are creating huge amounts of data on daily basis through their interactions with software. It has a huge population of relatively low-cost college graduates, for doing the more repetitive work of categorizing data. And it also has more top artificial-intelligence researchers than any country other than the US; indeed, many of the top Chinese in the field were educated in the US and have worked for US companies.

China also has a government that has decided that artificial intelligence is going to be the key technology of the future, and that will not accept being left behind. An ambitious national plan released in July 2017 calls for China to lead the world in artificial intelligence theory, technology and applications by 2030 (for detailed analysis of the plan, see these reports by the Paulson Institute and Oxford University’s Future of Humanity Institute). While it is difficult for government plans to create fundamental research breakthroughs on demand, such plans can be good at mobilizing lots of resources. So to the extent that advances in machine learning are now about mobilizing resources, it is reasonable to think China will indeed be able to make lots of progress.

AI-chart2

The prospect of China being something close to a peer of the US in a major new technology is a shocking development for many Americans. Everyone knows that China’s economy has grown rapidly and that it has accomplished a lot. But most of its past successes in technology involve deploying things developed elsewhere, such as mobile phones, wind turbines or high-speed trains, on a large scale. China has a per-capita GDP of roughly US$8-9,000 at market exchange rates, lower than Mexico or Turkey—and no one is talking about their dominance in the technologies of the future. It is tempting to try to resolve this paradox by focusing on China’s state support for artificial intelligence, implying that its advantages are unfair. The rivalry is also not a purely economic one, since there are military uses for machine learning.

The paradox is more apparent than real. China is such a huge, diverse and unequal country that averages are not a good guide to the location of the cutting edge. The reality, as anyone who has visited Beijing, Shanghai, or Shenzhen in recent years can attest, is that the income, skills and education levels of its best people can be comparable to those in the US. That elite of course is not representative of all the hundreds of millions of their compatriots, but neither is the Silicon Valley elite representative of middle-class Americans.

The fact that China now has the capability to contribute to cutting-edge research is also in large part a result of its integration with the US: it is the decades of sending top Chinese students to top US universities that have built up the necessary human capital. Rather than say there is a competition between the artificial intelligence sectors in the US and China, it might be more accurate to say that there is a single, global field of machine-learning research that has a significant presence in both North America (Canada also has some top people) and China.

There is no AI race

More fundamentally, it is wrong to think of China and the US as being in a “race” for supremacy in artificial intelligence. Evoking the “space race” with the Soviet Union in the 1960s is the wrong analogy. The space race was about achieving clear technical landmarks defined in advance: first satellite in orbit, first human in orbit, first human on the moon, etc. Today, it’s not clear what the technical landmarks for an artificial intelligence race might be. There is a vague goal of “general purpose artificial intelligence,” which means the kind of thinking, talking computers that are familiar from decades of portrayal in science fiction. But there is no race to make one, since no one knows how.

Rather, there are multiple related efforts going on to make progress on diverse sets of specific technical challenges and applications. If China is the first to achieve some technical breakthrough, that does not prevent the US from also doing so, nor does it guarantee that a Chinese company will control the market for applying that breakthrough. Recall that machine-learning applications can be “narrow” and “brittle”: software that is excellent at predicting, say, the video-watching habits of Chinese will not necessarily also dominate the American market. What we can say is that there are economies of scale and scope in machine-learning research: teams of experts who have successfully developed one machine-learning application themselves learn things that will make them better at developing other machine-learning applications (see this recent paper by Avi Goldfarb and Daniel Trefler for more).

Artificial intelligence is not a prize to be won, or even a single technology. Machine learning is a technique for solving problems and making software. At the moment, it is far from clear what the most commercially important use of machine learning will be. In a way, it is a solution in search of problems. China is making a big push in this area not because it knows what artificial intelligence will be able to do and wants to get there first, but because it does not know, and wants to make sure it does not lose out on the potential benefits. China’s development plan for artificial intelligence is mostly a laundry list of buzzwords and hoped-for technical breakthroughs.

The fact that machine learning is now in a resource-intensive phase does play to China’s strengths. There is an enormous amount of venture-capital money and government largesse flowing toward anything labeled “artificial intelligence” in China, and Chinese companies have had some notable successes in attracting high-profile figures in the field to join them. But fears that China will somehow monopolize the resources needed to make progress in machine learning are fanciful. After all, most of the key resources are human beings, who have minds of their own. And many of the key tools and concepts for creating machine-learning applications are in the public domain.

Will today’s advantages endure?

It is also not certain that the current resource-intensive phase of machine learning will last forever. As is usually the case when limited resources constrain development, people are trying to find ways to use fewer resources: in this case, refining machine learning so that it does not require so much human effort in categorizing data and fine-tuning algorithms. Some of the current buzzwords in the field are “unsupervised learning,” where the machine-learning algorithm is trained on raw data that is not classified or labeled, and “transfer learning,” where an algorithm that has already been trained on one dataset is repurposed onto another dataset, which requires much less data the second time around. Progress in these areas could lessen the advantages of China’s “big push” approach, though of course Chinese researchers would also benefit from them.

China’s current strength in machine learning is the result of a convergence between its own capabilities and the needs of the technology; since both are evolving, this convergence may not be a permanent one. But China’s government is correct to see the current moment as a great opportunity. China was already becoming one of the global clusters of machine-learning research even before the government decided to throw lots of subsidies at the technology. The self-reinforcing dynamics of clusters mean that today’s successes will make it easier for China to attract more machine-learning experts and companies in the future.

The biggest loser from this trend, however, is not the US, which already has well-established clusters of machine-learning research, but smaller nations who would also like to become home to such clusters. European countries, for instance, seem to be struggling to hold their own. The perception that there is a rivalry or race between the US and China ultimately derives from the fact that the two countries are rivals rather than friends. Artificial intelligence may indeed be the first example of a major cutting-edge technology whose development is led by geopolitical competitors—the US and China—rather than a group of friendly nations. The rising tensions between the US and China pose the question of whether a global artificial-intelligence field structured in this way is sustainable, or will be forced to split into national communities. The loss of those exchanges would slow progress in both countries.