Philippa Kaur

IF YOU LISTEN to the bombastic rhetoric in Beijing and Washington, America and China are engaged in an all-out contest for technological supremacy. “Fundamentally, we believe that a select few technologies are set to play an outsized importance over the coming decade,” thundered Jake Sullivan, President Joe Biden’s national security adviser, last September. In February Xi Jinping, China’s paramount leader, echoed the sentiment, stating that “we urgently need to strengthen basic research and solve key technology problems” in order to “cope with international science and technology competition, achieve a high level of self-reliance and self-improvement”. No technology seems to obsess policymakers on both sides of the Pacific more right now than artificial intelligence (AI). The rapid improvements in the abilities of “generative” AIs like ChatGPT, which analyse the web’s worth of human text, images or sounds and can then create increasingly passable simulacrums, have only strengthened the obsession. If generative AI proves as transformational as its boosters claim, the technology could give those who wield them an economic and military edge in the 21st century’s chief geopolitical contest. Western and Chinese strategists already talk of an AI arms race. Can China win it?On some measures of AI prowess, the autocracy pulled ahead some time ago (see chart 1). China surpassed America in the share of highly cited AI papers in 2019; in 2021 26% of AI conference publications globally came from China, compared with America’s share of 17%. Nine of the world’s top ten institutions, by volume of AI publications, are Chinese. According to one popular benchmark, so are the top five labs working on computer vision, a type of AI particularly useful to a communist surveillance state.At the same time, when it comes to “foundation models”, which give the buzzy generative AIs their wits, America finds itself firmly in front (see chart 2). ChatGPT and the pioneering model behind it, the latest version of which is called GPT-4, are the brain child of OpenAI, an American startup. A handful of other American firms, from small firms such as Anthropic or Stability AI to tech giants like Google, Meta and Microsoft (which part-owns OpenAI), have their own powerful systems. ERNIE, a Chinese rival to ChatGPT built by Baidu, China’s internet-search giant, is widely seen as less clever than most of them (see chart 3). Alibaba and Tencent, China’s mightiest tech titans, have yet even to unveil their own generative AIs. This leads those in the know to conclude that China is two or three years behind America in foundation-model building. There are three reasons for this underperformance. The first concerns data. On the surface, a centralised autocracy should be able to marshal a lot of it—the government was, for instance, able to hand over its troves of surveillance information on Chinese citizens to companies such as SenseTime or Megvii that, with the help of the country’s leading computer-vision labs, then used it to develop top-notch facial-recognition systems. That advantage has proved less formidable in the context of generative AIs, because foundation models are trained on the much more voluminous unstructured data of the internet. American model-builders benefit from the fact that 56% of all websites are in English, whereas just 1.5% are in Mandarin or China’s other languages, according to data from the W3Techs, an internet-research site. As Yiqin Fu of Stanford University points out, the Chinese interact with the internet primarily through mobile super-apps like WeChat and Weibo. These are “walled gardens”, so much of their content is not indexed on search engines. This makes that content harder for AI models to suck up. Lack of data may explain why Wu Dao 2.0, a Chinese model unveiled in 2021 by the Beijing Academy of Artificial Intelligence, a state-backed outfit, failed to make a splash despite its possibly being more computationally complex than GPT-4.The second reason for China’s lacklustre generative achievements has to do with hardware. Last year America imposed swingeing export controls on any technology that might give its main geostrategic rival a leg-up in AI. In particular, that includes the powerful chips used in the cloud-computing data centres where foundation models do their learning, and the chipmaking tools that could enable China to build such semiconductors on its own. That was a blow to Chinese model-builders. An analysis of 26 big Chinese models by the Centre for the Governance of AI, a British think-tank, found that more than half depended on Nvidia, an American chip designer, for their processing power. Some reports suggest that SMIC, China’s biggest chip manufacturer, has produced prototype chips which are just a generation or two behind TSMC, the Taiwanese industry leader that manufactures chips for Nvidia (see chart 4). But the Chinese firm may only be able to mass-produce chips which TSMC was churning out by the million three or four years ago. A professor at a leading Chinese university laments his country’s weakness in such “basic infrastructure” of AI.Chinese AI firms are also having more trouble getting their hands on another American export: know-how. America remains a magnet for the world’s tech talent; two-thirds of AI experts in America who present papers at the biggest AI conference are foreign-born. Chinese engineers made up 27% of that select group in 2019. Many Chinese AI boffins studied or worked in America before bringing their machine learnings back home. (Few non-Chinese boffins would consider moving to a police state a wise career move.) The covid-19 pandemic and rising Sino-American tensions are causing their numbers to dwindle. In the first half of 2022 America granted half as many visas to Chinese students as in the same period in 2019. The triple shortage—of data, hardware and expertise—has been a genuine hurdle for China. Whether it will hold Chinese AI ambitions back much longer is, though, another matter. Info attainmentTake data. On February 13th the local authorities in Beijing, where nearly a third of China’s AI firms are located, said they were releasing data from 115 state-affiliated organisations, giving model-builders 15,880 data sets to play with. To liberate more data, the central government also wants to dismantle Chinese apps’ walled gardens. Most important, the latest models appear able to transfer learnings from one language to another. In the paper describing GPT-4, OpenAI said that the model performed remarkably well on tasks in Chinese despite the dearth of Chinese source material in the model’s training data. Already Baidu’s ERNIE was trained on lots of English-language data, notes Jeffrey Ding of George Washington University. In hardware, too, China is finding workarounds. The Financial Times reported in March that SenseTime, which is blacklisted by America’s government, has used intermediaries to skirt the export controls. Some Chinese AI firms are able to harness the computing power of Nvidia’s advanced chips through cloud servers based in other countries. Alternatively, they can simply buy more of Nvidia’s less advanced semiconductors or use them more efficiently with the help of clever software. To continue serving the vast Chinese market, the American company has designed less powerful sanctions-compliant processors. These are between 10% and 30% slower than its top-of-the-range kit, and end up being costlier for the Chinese customers per unit of processing power. But they do the job. China could partly alleviate the dearth of chips—and of brain power—with the help of “open-source” models. Such models’ inner workings can be downloaded by anyone and fine-tuned to a specific task. Most importantly, that includes the numbers, called “weights”, which define the structure of the model and which are derived from costly training runs. Alpaca, a model built by researchers at Stanford University using the weights from LLaMA, a foundation model created by Meta, was made for less than $600, compared with sums on the order of $100m for training something like GPT-4. Alpaca performs just as well as the original version of ChatGPT on many tasks. Chinese AI labs could similarly avail themselves of open-source models, which embody the collective wisdom of international research teams. Matt Sheehan of the Carnegie Endowment for International Peace, another think-tank, says that China has form in being a “fast follower”—its labs have absorbed advances from abroad and then rapidly incorporated them into their own models, often with flush state resources. A prominent Silicon Valley venture capitalist is more blunt, calling open-source models a gift to the Communist Party.Such considerations make it hard to imagine that either America or China could in the long run build an unbridgeable lead in AI modelling. Each may well end up with AIs of roughly similar ability, even if it costs China over the odds to keep up in the face of American sanctions. But even if the race of the model-builders is a dead heat, America has one thing going for it that could make it the big AI winner—its peerless ability to spread cutting-edge innovation throughout the economy. It was, after all, more efficient diffusion of technology that helped America open up a technological lead over the Soviet Union, which in the 1950s was producing twice as many science PhDs as its democratic adversary.China is, of course, far more competent than the Soviet Union ever was at adopting new technologies. Its fintech platforms, 5G telecoms and high-speed rail are all world-class. But those successes may be the exception, not the rule, says Mr Ding. Particularly, in the deployment of sensors, cloud computing and business software—all complementary to AI—China has done less well. Although American export controls may not derail all Chinese model-building, they do constrain China’s tech industry more broadly, thereby slowing the adoption of any new technology. Moreover, corporate China as a whole, and especially small and medium-sized companies, is short of technologists who act as conduits for technological diffusion. Swathes of the economy are dominated by state-owned firms, which tend to be stodgy and change-averse. China’s “Big Fund” for chips, which raised $50bn in 2014 with a view to backing domestic semiconductor firms, has been mired in scandals. Many of the thousands of AI startups created in recent years have simply slapped on the AI label in the hope of getting a slice of the lavish subsidies doled out by the state to the favoured industry. As a consequence, China’s private sector may find it hard to take full advantage of generative AI, especially if the Communist Party imposes heavy regulations to prevent chatbots from saying something its censors do not like. Such handicaps would come on top of Mr Xi’s broader suborning of private enterprise, including a two-and-a-half-year crackdown on China’s tech industry. Although this anti-tech campaign has officially ended, it has left businesses scarred. The result is a chill in tech sentiment. Last year private investments in Chinese AI startups amounted to $13.5bn, less than one-third the amount that flowed to their American rivals. In the first four months of 2023 the funding gap appears only to have widened, according to PitchBook, a data provider. Whether or not generative AI proves revolutionary, the free market has placed its bet on who will make the most of it. ■ More

IF YOU LISTEN to the bombastic rhetoric in Beijing and Washington, America and China are engaged in an all-out contest for technological supremacy. “Fundamentally, we believe that a select few technologies are set to play an outsized importance over the coming decade,” thundered Jake Sullivan, President Joe Biden’s national security adviser, last September. In February Xi Jinping, China’s paramount leader, echoed the sentiment, stating that “we urgently need to strengthen basic research and solve key technology problems” in order to “cope with international science and technology competition, achieve a high level of self-reliance and self-improvement”. No technology seems to obsess policymakers on both sides of the Pacific more right now than artificial intelligence (AI). The rapid improvements in the abilities of “generative” AIs like ChatGPT, which analyse the web’s worth of human text, images or sounds and can then create increasingly passable simulacrums, have only strengthened the obsession. If generative AI proves as transformational as its boosters claim, the technology could give those who wield them an economic and military edge in the 21st century’s chief geopolitical contest. Western and Chinese strategists already talk of an AI arms race. Can China win it?On some measures of AI prowess, the autocracy pulled ahead some time ago (see chart 1). China surpassed America in the share of highly cited AI papers in 2019; in 2021 26% of AI conference publications globally came from China, compared with America’s share of 17%. Nine of the world’s top ten institutions, by volume of AI publications, are Chinese. According to one popular benchmark, so are the top five labs working on computer vision, a type of AI particularly useful to a communist surveillance state.At the same time, when it comes to “foundation models”, which give the buzzy generative AIs their wits, America finds itself firmly in front (see chart 2). ChatGPT and the pioneering model behind it, the latest version of which is called GPT-4, are the brain child of OpenAI, an American startup. A handful of other American firms, from small firms such as Anthropic or Stability AI to tech giants like Google, Meta and Microsoft (which part-owns OpenAI), have their own powerful systems. ERNIE, a Chinese rival to ChatGPT built by Baidu, China’s internet-search giant, is widely seen as less clever than most of them (see chart 3). Alibaba and Tencent, China’s mightiest tech titans, have yet even to unveil their own generative AIs. This leads those in the know to conclude that China is two or three years behind America in foundation-model building. There are three reasons for this underperformance. The first concerns data. On the surface, a centralised autocracy should be able to marshal a lot of it—the government was, for instance, able to hand over its troves of surveillance information on Chinese citizens to companies such as SenseTime or Megvii that, with the help of the country’s leading computer-vision labs, then used it to develop top-notch facial-recognition systems. That advantage has proved less formidable in the context of generative AIs, because foundation models are trained on the much more voluminous unstructured data of the internet. American model-builders benefit from the fact that 56% of all websites are in English, whereas just 1.5% are in Mandarin or China’s other languages, according to data from the W3Techs, an internet-research site. As Yiqin Fu of Stanford University points out, the Chinese interact with the internet primarily through mobile super-apps like WeChat and Weibo. These are “walled gardens”, so much of their content is not indexed on search engines. This makes that content harder for AI models to suck up. Lack of data may explain why Wu Dao 2.0, a Chinese model unveiled in 2021 by the Beijing Academy of Artificial Intelligence, a state-backed outfit, failed to make a splash despite its possibly being more computationally complex than GPT-4.The second reason for China’s lacklustre generative achievements has to do with hardware. Last year America imposed swingeing export controls on any technology that might give its main geostrategic rival a leg-up in AI. In particular, that includes the powerful chips used in the cloud-computing data centres where foundation models do their learning, and the chipmaking tools that could enable China to build such semiconductors on its own. That was a blow to Chinese model-builders. An analysis of 26 big Chinese models by the Centre for the Governance of AI, a British think-tank, found that more than half depended on Nvidia, an American chip designer, for their processing power. Some reports suggest that SMIC, China’s biggest chip manufacturer, has produced prototype chips which are just a generation or two behind TSMC, the Taiwanese industry leader that manufactures chips for Nvidia (see chart 4). But the Chinese firm may only be able to mass-produce chips which TSMC was churning out by the million three or four years ago. A professor at a leading Chinese university laments his country’s weakness in such “basic infrastructure” of AI.Chinese AI firms are also having more trouble getting their hands on another American export: know-how. America remains a magnet for the world’s tech talent; two-thirds of AI experts in America who present papers at the biggest AI conference are foreign-born. Chinese engineers made up 27% of that select group in 2019. Many Chinese AI boffins studied or worked in America before bringing their machine learnings back home. (Few non-Chinese boffins would consider moving to a police state a wise career move.) The covid-19 pandemic and rising Sino-American tensions are causing their numbers to dwindle. In the first half of 2022 America granted half as many visas to Chinese students as in the same period in 2019. The triple shortage—of data, hardware and expertise—has been a genuine hurdle for China. Whether it will hold Chinese AI ambitions back much longer is, though, another matter. Info attainmentTake data. On February 13th the local authorities in Beijing, where nearly a third of China’s AI firms are located, said they were releasing data from 115 state-affiliated organisations, giving model-builders 15,880 data sets to play with. The central government has previously signalled it wants to dismantle Chinese apps’ walled gardens, potentially liberating more data, says Kayla Blomquist, a former American diplomat in China now at Oxford University.Most important, the latest models appear able to transfer learnings from one language to another. In the paper describing GPT-4, OpenAI said that the model performed remarkably well on tasks in Chinese despite the dearth of Chinese source material in the model’s training data. Already Baidu’s ERNIE was trained on lots of English-language data, notes Jeffrey Ding of George Washington University. In hardware, too, China is finding workarounds. The Financial Times reported in March that SenseTime, which is blacklisted by America’s government, has used intermediaries to skirt the export controls. Some Chinese AI firms are able to harness the computing power of Nvidia’s advanced chips through cloud servers based in other countries. Alternatively, they can simply buy more of Nvidia’s less advanced semiconductors or use them more efficiently with the help of clever software. To continue serving the vast Chinese market, the American company has designed less powerful sanctions-compliant processors. These are between 10% and 30% slower than its top-of-the-range kit, and end up being costlier for the Chinese customers per unit of processing power. But they do the job. China could partly alleviate the dearth of chips—and of brain power—with the help of “open-source” models. Such models’ inner workings can be downloaded by anyone and fine-tuned to a specific task. Most importantly, that includes the numbers, called “weights”, which define the structure of the model and which are derived from costly training runs. Alpaca, a model built by researchers at Stanford University using the weights from LLaMA, a foundation model created by Meta, was made for less than $600, compared with sums on the order of $100m for training something like GPT-4. Alpaca performs just as well as the original version of ChatGPT on many tasks. Chinese AI labs could similarly avail themselves of open-source models, which embody the collective wisdom of international research teams. Matt Sheehan of the Carnegie Endowment for International Peace, another think-tank, says that China has form in being a “fast follower”—its labs have absorbed advances from abroad and then rapidly incorporated them into their own models, often with flush state resources. A prominent Silicon Valley venture capitalist is more blunt, calling open-source models a gift to the Communist Party.Such considerations make it hard to imagine that either America or China could in the long run build an unbridgeable lead in AI modelling. Each may well end up with AIs of roughly similar ability, even if it costs China over the odds to keep up in the face of American sanctions. But even if the race of the model-builders is a dead heat, America has one thing going for it that could make it the big AI winner—its peerless ability to spread cutting-edge innovation throughout the economy. It was, after all, more efficient diffusion of technology that helped America open up a technological lead over the Soviet Union, which in the 1950s was producing twice as many science PhDs as its democratic adversary.China is, of course, far more competent than the Soviet Union ever was at adopting new technologies. Its fintech platforms, 5G telecoms and high-speed rail are all world-class. But those successes may be the exception, not the rule, says Mr Ding. Particularly in the deployment of sensors, cloud computing and business software—all complementary to AI—China has done less well. Although American export controls may not derail all Chinese model-building, they do constrain China’s tech industry more broadly, thereby slowing the adoption of any new technology. Moreover, corporate China as a whole, and especially small and medium-sized companies, is short of technologists who act as conduits for technological diffusion. Swathes of the economy are dominated by state-owned firms, which tend to be stodgy and change-averse. China’s “Big Fund” for chips, which raised $50bn in 2014 with a view to backing domestic semiconductor firms, has been mired in scandals. Many of the thousands of AI startups created in recent years have simply slapped on the AI label in the hope of getting a slice of the lavish subsidies doled out by the state to the favoured industry. As a consequence, China’s private sector may find it hard to take full advantage of generative AI, especially if the Communist Party imposes heavy regulations to prevent chatbots from saying something its censors do not like. Such handicaps would come on top of Mr Xi’s broader suborning of private enterprise, including a two-and-a-half-year crackdown on China’s tech industry. Although this anti-tech campaign has officially ended, it has left businesses scarred. The result is a chill in tech sentiment. Last year private investments in Chinese AI startups amounted to $13.5bn, less than one-third the amount that flowed to their American rivals. In the first four months of 2023 the funding gap appears only to have widened, according to PitchBook, a data provider. Whether or not generative AI proves revolutionary, the free market has placed its bet on who will make the most of it. ■ More

For a public demonstration of the importance of ritual, the coronation of King Charles III on May 6th will be hard to beat. The ceremony will take place at Westminster Abbey, where monarchs have been crowned since William the Conqueror in 1066. There will be anointing, homage-paying, oath-taking and all manner of processing. In any other circumstances this kind of behaviour would warrant a medical diagnosis. But the alchemy of tradition means that it will instead call forth a sense of continuity and the idea of shared history. Listen to this story. Enjoy more audio and podcasts on More

BEFORE CARL ICAHN was an activist investor, he was an arbitrageur. Although it was swashbuckling corporate raids during the 1980s that made him infamous, some of Mr Icahn’s earliest campaigns involved investing in closed-end funds, a type of investment company which often trades at a discount to the value of its assets. Closing this gap, perhaps by agitating for the fund to liquidate its holdings, yields a profit.Mr Icahn’s own investment holding company, Icahn Enterprises, suffered no such discount. Until this week the firm had a market capitalisation of around $18bn, more than triple the reported net value of its assets. These include majority ownership of energy and car companies, in addition to an activist-investment portfolio. On May 2nd Hindenburg Research, a short-selling outfit founded in 2017 by Nathan Anderson, accused Icahn Enterprises of operating a “Ponzi-like” structure. Icahn Enterprises has shed more than a third of its market value since Hindenburg released its report. It has become the latest of Hindenburg’s targets to hit the skids—and the headlines. Mr Anderson’s firm has previously taken aim at Nikola, a maker of electric lorries, the Adani Group, one of India’s mightiest conglomerates, and Block, an American fintech giant (see chart).Hindenburg’s latest report alleges that Icahn Enterprises has inflated the value of its assets and funded its dividend with proceeds from selling shares to unwitting investors. It also calls on Mr Icahn to disclose the terms of personal loans secured against his majority holding in Icahn Enterprises. And it scolds Jefferies, Mr Icahn’s long-time investment bankers and the only big bank whose research analysts cover Icahn Enterprises, for allegedly turning a blind eye to the firm’s risks. Mr Icahn, Hindenburg argues, “has made a classic mistake of taking on too much leverage in the face of sustained losses”. Bill Ackman, another famed activist investor who once locked horns with Mr Icahn over an investment in Herbalife, an American supplement firm, gloated on Twitter that there was a “karmic quality” to the report. Short-sellers’ targets can be hamstrung in their immediate defences—share prices can tank quickly but detailed rebuttals take time. Even so, Mr Icahn’s first response looks muted compared with that of Hindenburg’s recent victims. In March Block described Hindenburg’s report as “factually inaccurate” and threatened litigation. In January the Adani Group accused the short-seller of “selective misinformation”. After stating that Hindenburg’s report is “self-serving”, Mr Icahn said on May 2nd merely that his firm’s performance would “speak for itself”. Jefferies has not commented on Hindenburg’s claims. Quite how messy this activist showdown becomes remains to be seen. Hindenburg’s report pitches a doyen of classic shareholder activism, which involves trying to drive a target’s share price up, against a newly prominent practitioner of short-selling, which aims to send it through the floor. The stakes are higher for Mr Icahn. His brand of activism requires investors to take him more seriously than they do the bad managers that, in his “anti-Darwinian” view, American commerce seems to promote. Icahn Enterprises must now prove that the same thing is not true of its own boardroom. ■To stay on top of the biggest stories in business and technology, sign up to the Bottom Line, our weekly subscriber-only newsletter. More

AS CHRISTOPHER WAS preparing to board a flight from New York to Singapore in February 2019, he was pulled aside by local authorities and told to stay put. An Interpol “red notice”, a request for local law enforcement to make an arrest on behalf of another government, had been issued on his name, he would soon learn. The executive, who has asked that his real name not be used because his case is ongoing, was the founder of an international advertising group that a few years earlier had got into big trouble in China over data security. Listen to this story. Enjoy more audio and podcasts on More

A sensational scoop was tweeted last month by America’s National Public Radio: Elon Musk’s “massive space sex rocket” had exploded on launch. Alas, it turned out to be an automated mistranscription of SpaceX, the billionaire’s rocketry firm. The error may be a taste of what is to come as artificial intelligence (AI) plays a bigger role in newsrooms.Listen to this story. Enjoy more audio and podcasts on More

Two work laptops, two work calendars, two bosses and two pay-cheques. So far, neither of Matt’s employers is any the wiser. The tech worker (who, for obvious reasons, asked The Economist not to use his real name) meets deadlines and does what is requested, though not more. He is not the only one.People working several jobs is nothing new. Low earners have long had to juggle shifts to make ends meet. At the other end of the pay scale, directors often sit on a few corporate boards. According to America’s Bureau of Labour Statistics, at any given point in the past 30 years, between 4% and 6.5% of the American workforce was working more than one job. Estimates from the Census Bureau put that share even higher, going from 6.8% in 1996 to 7.8% in 2018. What is novel, as Matt’s example illustrates, is the rise of the job-juggling white-collar type, especially in the technology industry. Thank—or blame—remote work. Despite efforts by bosses to lure or coerce people back to their desks, the share of techies working fully remotely remains 60% higher than in other sectors (see chart). Without managers physically looking over their shoulders, some of them are two-timing their employers. Mid-career software engineers report applying for more junior positions so that they can “underpromise and overdeliver”, with minimal effort. Matt took a second job, or “J2” as he calls it, for two main reasons: boredom and concerns over job security. The tasks required by his first job, working remotely as a data scientist for a medium-sized tech firm, were not particularly challenging, taking him only eight hours a week. He had no inclination to “play office politics and move up the corporate ladder”. He did, though, covet cash. He reckoned he could take on a second job, double his pay and gain a safety-net were he to be laid off.After interviewing for a few weeks, Matt found a promising J2: data engineering at a startup. He suspected that demands on his time would be as low as they were at his first job. He was mostly right, though striking a balance required some footwork. In his first week a rare J1 meeting was scheduled at the same time as one of his J2 “onboarding” sessions. Some fellow members of an online forum for the overemployed on Reddit, a social-media site, claim to have taken two meetings at once, with video off. If called on to speak at the same time, they feign connectivity problems or play a pre-recorded audio clip of a dog barking. Matt decided to tune in to the J1 call and reschedule his onboarding, blaming a doctor’s appointment.The rise of generative artificial intelligence like ChatGPT may in time make double-jobbing harder by replacing some menial tech tasks. Until then, coasters can themselves use clever chatbots to help structure computer code, write documents and even conduct preliminary research. ChatGPT cannot replace the work of a software engineer, says one overemployee, but it gets you 90% of the way there. The employee-employer relationship has historically favoured the employers, who wield more clout because they can typically choose from more workers than workers can among companies. Matt thinks of his ruse as taking back some control. Two decently paying jobs afford him flexibility. And, he says, flexibility is power. If he were to get laid off, or if one job were to become unreasonably demanding, he could go and find another. For now, he thinks he is safe. So safe, in fact, that he is starting his search for a third job. ■To stay on top of the biggest stories in business and technology, sign up to the Bottom Line, our weekly subscriber-only newsletter. More

IN MARCH AMAZON announced it would fire 9,000 workers—bringing to 27,000 the total number it has laid off this year. The e-commerce giant’s share price is down by a third since 2021. Other online-shopping darlings, from Shopify in Canada to Coupang in South Korea and Grab in South-East Asia, have suffered a similar fate (see chart). With one exception. At $64bn, the market value of MercadoLibre, an Argentine firm listed in New York with operations across Latin America, has been rising lately and is back roughly to where it was at the start of 2022—and twice that before covid-19. In April, as the world’s tech firms were sacking workers en masse, it said it would hire 13,000, mainly in Brazil and Mexico, raising its workforce by a third. MercadoLibre needs more workers. On May 3rd it reported that revenues grew by 35% in the first quarter of 2023, to $3bn. Last year goods worth $35bn changed hands on its platform, helping generate $1bn in pre-tax profits. How is it flourishing as similar firms elsewhere struggle?Its success is a mix of good management and good fortune. Early on it expanded from connecting buyers and sellers into payments, initially to allay users’ fear of fraud. Its payments system, MercadoPago, is now widely trusted and used beyond its platform; more than $100bn flowed through it in 2022. The company has also built its own logistics network to deliver packages quickly in a region where infrastructure can be patchy. In ten years it has gone from not touching parcels, all of which were handled by third-party shippers, to having a hand in ferrying 93% of its e-commerce packages. More recently it added a fast-growing advertising business. Unlike Amazon, which regularly receives complaints about working conditions, employees rank MercadoLibre among the best Latin American firms to work for. MercadoLibre also benefits from a deep understanding of local shopping habits, notes Ricardo Tapia of the University of Anáhuac in Mexico City. For instance, by accumulating points for purchases, its shoppers can gain benefits such as free delivery. What may seem gimmicky to Western shoppers, for whom a big benefit of buying online is that it saves time, is a big draw for game-loving Latin Americans. The resulting strength has allowed the firm to take advantage of fortuitous circumstances. As everywhere in the world, the pandemic accelerated the growth of e-commerce in its region. In Mexico, MercadoLibre’s third-biggest market after Brazil and Argentina, 63m people bought something online in 2022, up from 37m in 2018. In contrast to more mature markets such as Britain, the number of Latin Americans buying online did not drop back down after an initial boom in 2020. The region’s brick-and-mortar retailers, which are rapidly improving their own digital offerings, and online giants such as Amazon have cottoned on to this trend. To keep growing, MercadoLibre may need to boost penetration in less online countries such as Colombia, where Amazon is weaker, and perhaps move into new segments, such as groceries. But it does at least enjoy another advantage over foreign rivals, for which Latin America is a peripheral market—focus. Failure in its home region is simply not an option, says Agustin Gutierrez of McKinsey, a consultancy. Nothing concentrates the mind like survival. ■To stay on top of the biggest stories in business and technology, sign up to the Bottom Line, our weekly subscriber-only newsletter. More