The quest to type Chinese on a QWERTY keyboard created autocomplete

This is an excerpt from The Chinese Computer: A Global History of the Information Age by Thomas S. Mullaney, published on May 28 by The MIT Press. It has been lightly edited.

ymiw2

klt4

pwyy1

wdy6

o1

dfb2

wdv2

fypw3

uet5

dm2

dlu1 …

A young Chinese man sat down at his QWERTY keyboard and rattled off an enigmatic string of letters and numbers.

Was it code? Child’s play? Confusion? It was Chinese.

The beginning of Chinese, at least. These forty-four keystrokes marked the first steps in a process known as “input” or shuru: the act of getting Chinese characters to appear on a computer monitor or other digital device using a QWERTY keyboard or trackpad.

COURTESY OF MIT PRESS

Across all computational and digital media, Chinese text entry relies on software programs known as “Input Method Editors”—better known as “IMEs” or simply “input methods” (shurufa). IMEs are a form of “middleware,” so-named because they operate in between the hardware of the user’s device and the software of its program or application. Whether a person is composing a Chinese document in Microsoft Word, searching the web, sending text messages, or otherwise, an IME is always at work, intercepting all of the user’s keystrokes and trying to figure out which Chinese characters the user wants to produce. Input, simply put, is the way ymiw2klt4pwyy … becomes a string of Chinese characters.

IMEs are restless creatures. From the moment a key is depressed, or a stroke swiped, they set off on a dynamic, iterative process, snatching up user-inputted data and searching computer memory for potential Chinese character matches. The most popular IMEs these days are based on Chinese phonetics—that is, they use the letters of the Latin alphabet to describe the sound of Chinese characters, with mainland Chinese operators using the country’s official Romanization system, Hanyu pinyin. 

COURTESY OF MIT PRESS

This young man’s name was Huang Zhenyu (also known by his nom de guerre, Yu Shi). He was one of around sixty contestants that day, each wearing a bright red shoulder sash—like a tickertape parade of old, or a beauty pageant. “Love Chinese Characters” (Ai Hanzi) was emblazoned in vivid, golden yellow on a poster at the front of the hall. The contestants’ task was to transcribe a speech by outgoing Chinese president Hu Jintao, as quickly and as accurately as they could. “Hold High the Great Banner of Socialism with Chinese Characteristics,” it began, or in the original:  高举中国特色社会主义伟大旗帜为夺取全面建设小康社会新胜利而奋斗. Huang’s QWERTY keyboard did not permit him to enter these characters directly, however, and so he entered the quasi-gibberish string of letters and numbers instead: ymiw2klt4pwyy1wdy6…

With these four-dozen keystrokes, Huang was well on his way, not only to winning the 2013 National Chinese Characters Typing Competition, but also to clock one of the fastest typing speeds ever recorded, anywhere in the world.

ymiw2klt4pwyy1wdy6 … is not the same as 高举中国特色社会主义 …  the keys that Huang actually depressed on his QWERTY keyboard—his “primary transcript,” as we could call it—were completely different than the symbols that ultimately appeared on his computer screen, namely the “secondary transcript” of Hu Jintao’s speech. This is true for every one of the world’s billion-plus Sinophone computer users. In Chinese computing, what you type is never what you get.

For readers accustomed to English-language word processing and computing, this should come as a surprise. For example, were you to compare the paragraph you’re reading right now against a key log showing exactly which buttons I depressed to produce it, the exercise would be unenlightening (to put it mildly). “F-o-r-_-r-e-a-d-e-r-s-_-a-c-c-u-s-t-o-m-e-d-_t-o-_-E-n-g-l-i-s-h … ” it would read (forgiving any typos or edits). In English-language typewriting and computer input, a typist’s primary and secondary transcripts are, in principle, identical. The symbols on the keys and the symbols on the screen are the same.

Not so for Chinese computing. When inputting Chinese, the symbols a person sees on their QWERTY keyboard are always different from the symbols that ultimately appear on the monitor or on paper. Every single computer and new media user in the Sinophone world—no matter if they are blazing-fast or molasses-slow—uses their device in exactly the same way as Huang Zhenyu, constantly engaged in this iterative process of criteria-candidacy-confirmation, using one IME or another. Not some Chinese-speaking users, mind you, but all. This is the first and most basic feature of Chinese computing: Chinese human-computer interaction (HCI) requires users to operate entirely in code all the time.

If Huang Zhenyu’s mastery of a complex alphanumeric code weren’t impressive enough, consider the staggering speed of his performance. He transcribed the first 31 Chinese characters of Hu Jintao’s speech in roughly 5 seconds, for an extrapolated speed of 372 Chinese characters per minute. By the close of the grueling 20-minute contest, one extending over thousands of characters, he crossed the finish line with an almost unbelievable speed of 221.9 characters per minute.

That’s 3.7 Chinese characters every second.

In the context of English, Huang’s opening 5 seconds would have been the equivalent of around 375 English words-per-minute, with his overall competition speed easily surpassing 200 WPM—a blistering pace unmatched by anyone in the Anglophone world (using QWERTY, at least). In 1985, Barbara Blackburn achieved a Guinness Book of World Records–verified performance of 170 English words-per-minute (on a typewriter, no less). Speed demon Sean Wrona later bested Blackburn’s score with a performance of 174 WPM (on a computer keyboard, it should be noted). As impressive as these milestones are, the fact remains: had Huang’s performance taken place in the Anglophone world, it would be his name enshrined in the Guinness Book of World Records as the new benchmark to beat.

Huang’s speed carried special historical significance as well.

For a person living between the years 1850 and 1950—the period examined in the book The Chinese Typewriter—the idea of producing Chinese by mechanical means at a rate of over two hundred characters per minute would have been virtually unimaginable. Throughout the history of Chinese telegraphy, dating back to the 1870s, operators maxed out at perhaps a few dozen characters per minute. In the heyday of mechanical Chinese typewriting, from the 1920s to the 1970s, the fastest speeds on record were just shy of eighty characters per minute (with the majority of typists operating at far slower rates). When it came to modern information technologies, that is to say, Chinese was consistently one of the slowest writing systems in the world.

What changed? How did a script so long disparaged as cumbersome and helplessly complex suddenly rival—exceed, even—computational typing speeds clocked in other parts of the world? Even if we accept that Chinese computer users are somehow able to engage in “real time” coding, shouldn’t Chinese IMEs result in a lower overall “ceiling” for Chinese text processing as compared to English? Chinese computer users have to jump through so many more hoops, after all, over the course of a cumbersome, multistep process: the IME has to intercept a user’s keystrokes, search in memory for a match, present potential candidates, and wait for the user’s confirmation. Meanwhile, English-language computer users need only depress whichever key they wish to see printed on screen. What could be simpler than the “immediacy” of “Q equals Q,” “W equals W,” and so on?

COURTESY OF TOM MULLANEY

To unravel this seeming paradox, we will examine the first Chinese computer ever designed: the Sinotype, also known as the Ideographic Composing Machine. Debuted in 1959 by MIT professor Samuel Hawks Caldwell and the Graphic Arts Research Foundation, this machine featured a QWERTY keyboard, which the operator used to input—not the phonetic values of Chinese characters—but the brushstrokes out of which Chinese characters are composed. The objective of Sinotype was not to “build up” Chinese characters on the page, though, the way a user builds up English words through the successive addition of letters. Instead, each stroke “spelling” served as an electronic address that Sinotype’s logical circuit used to retrieve a Chinese character from memory. In other words, the first Chinese computer in history was premised on the same kind of “additional steps” as seen in Huang Zhenyu’s prizewinning 2013 performance.

During Caldwell’s research, he discovered unexpected benefits of all these additional steps—benefits entirely unheard of in the context of Anglophone human-machine interaction at that time. The Sinotype, he found, needed far fewer keystrokes to find a Chinese character in memory than to compose one through conventional means of inscription. By way of analogy, to “spell” a nine-letter word like “crocodile” (c-r-o-c-o-d-i-l-e) took far more time than to retrieve that same word from memory (“c-r-o-c-o-d” would be enough for a computer to make an unambiguous match, after all, given the absence of other words with similar or identical spellings). Caldwell called his discovery “minimum spelling,” making it a core part of the first Chinese computer ever built. 

Today, we know this technique by a different name: “autocompletion,” a strategy of human-computer interaction in which additional layers of mediation result in faster textual input than the “unmediated” act of typing. Decades before its rediscovery in the Anglophone world, then, autocompletion was first invented in the arena of Chinese computing.

Deepfakes of your dead loved ones are a booming Chinese business

Once a week, Sun Kai has a video call with his mother. He opens up about work, the pressures he faces as a middle-aged man, and thoughts that he doesn’t even discuss with his wife. His mother will occasionally make a comment, like telling him to take care of himself—he’s her only child. But mostly, she just listens.

That’s because Sun’s mother died five years ago. And the person he’s talking to isn’t actually a person, but a digital replica he made of her—a moving image that can conduct basic conversations. They’ve been talking for a few years now. 

After she died of a sudden illness in 2019, Sun wanted to find a way to keep their connection alive. So he turned to a team at Silicon Intelligence, an AI company based in Nanjing, China, that he cofounded in 2017. He provided them with a photo of her and some audio clips from their WeChat conversations. While the company was mostly focused on audio generation, the staff spent four months researching synthetic tools and generated an avatar with the data Sun provided. Then he was able to see and talk to a digital version of his mom via an app on his phone. 

“My mom didn’t seem very natural, but I still heard the words that she often said: ‘Have you eaten yet?’” Sun recalls of the first interaction. Because generative AI was a nascent technology at the time, the replica of his mom can say only a few pre-written lines. But Sun says that’s what she was like anyway. “She would always repeat those questions over and over again, and it made me very emotional when I heard it,” he says.

There are plenty of people like Sun who want to use AI to preserve, animate, and interact with lost loved ones as they mourn and try to heal. The market is particularly strong in China, where at least half a dozen companies are now offering such technologies and thousands of people have already paid for them. In fact, the avatars are the newest manifestation of a cultural tradition: Chinese people have always taken solace from confiding in the dead. 

The technology isn’t perfect—avatars can still be stiff and robotic—but it’s maturing, and more tools are becoming available through more companies. In turn, the price of “resurrecting” someone—also called creating “digital immortality” in the Chinese industry—has dropped significantly. Now this technology is becoming accessible to the general public. 

Some people question whether interacting with AI replicas of the dead is actually a healthy way to process grief, and it’s not entirely clear what the legal and ethical implications of this technology may be. For now, the idea still makes a lot of people uncomfortable. But as Silicon Intelligence’s other cofounder, CEO Sima Huapeng, says, “Even if only 1% of Chinese people can accept [AI cloning of the dead], that’s still a huge market.” 

AI resurrection

Avatars of the dead are essentially deepfakes: the technologies used to replicate a living person and a dead person aren’t inherently different. Diffusion models generate a realistic avatar that can move and speak. Large language models can be attached to generate conversations. The more data these models ingest about someone’s life—including photos, videos, audio recordings, and texts—the more closely the result will mimic that person, whether dead or alive.

China has proved to be a ripe market for all kinds of digital doubles. For example, the country has a robust e-commerce sector, and consumer brands hire many livestreamers to sell products. Initially, these were real people—but as MIT Technology Review reported last fall—many brands are switching to AI-cloned influencers that can stream 24/7. 

In just the past three years, the Chinese sector developing AI avatars has matured rapidly, says Shen Yang, a professor studying AI and media at Tsinghua University in Beijing, and replicas have improved from minutes-long rendered videos to 3D “live” avatars that can interact with people.  

This year, Sima says, has seen a tipping point, with AI cloning becoming affordable for most individuals. “Last year, it cost about $2,000 to $3,000, but it now only costs a few hundred dollars,” he says. That’s thanks to a price war between Chinese AI companies, which are fighting to meet the thriving demand for digital avatars in other sectors like streaming.

In fact, demand for applications that re-create the dead has also boosted the capabilities of tools that digitally replicate the living. 

Silicon Intelligence offers both services. When Sun and Sima launched the company, they were focused on using text-to-speech technologies to create audio and then using those AI-generated voices in applications such as robocalls.

But after the company replicated Sun’s mother, it pivoted to generating realistic avatars. That decision turned the company into one of the leading Chinese players creating AI-powered influencers. 

Its technology has generated avatars for hundreds of thousands of TikTok-like videos and streaming channels, but Sima says more recently it’s seen around 1,000 clients use it to replicate someone who’s passed away. “We started our work on ‘resurrection’ in 2019 and 2020,” he says, but at first people were slow to accept it: “No one wanted to be the first adopters.” 

The quality of the avatars has improved, he says, which has boosted adoption. When the avatar looks increasingly lifelike and gives fewer out-of-character answers, it’s easier for users to treat it as their deceased family member. Plus, the idea is getting popularized through more depictions on Chinese TV. 

Now Silicon Intelligence offers the replication service for a price between several hundred and several thousand dollars. The most basic product comes as an interactive avatar in an app, and the options at the upper end of the range often involve more customization and better hardware components, such as a tablet or a display screen. There are at least a handful more Chinese companies working on the same technology.

A modern twist on tradition

The business in these deepfakes builds on China’s long cultural history of communicating with the dead. 

In Chinese homes, it’s common to put up a portrait of a deceased relative for a few years after the death. Zhang Zewei, founder of a Shanghai-based company called Super Brain, says he and his team wanted to revamp that tradition with an “AI photo frame.” They create avatars of deceased loved ones that are pre-loaded onto an Android tablet, which looks like a photo frame when standing up. Clients can choose a moving image that speaks words drawn from an offline database or from an LLM. 

“In its essence, it’s not much different from a traditional portrait, except that it’s interactive,” Zhang says.

Zhang says the company has made digital replicas for over 1,000 clients since March 2023 and charges $700 to $1,400, depending on the service purchased. The company plans to release an app-only product soon, so that users can access the avatars on their phones, and could further reduce the cost to around $140.

The purpose of his products, Zhang says, is therapeutic. “When you really miss someone or need consolation during certain holidays, you can talk to the artificial living and heal your inner wounds,” he says.

And even if that conversation is largely one-sided, that’s in keeping with a strong cultural tradition. Every April during the Qingming festival, Chinese people sweep the tombs of their ancestors, burn joss sticks and fake paper money, and tell them what has happened in the past year. Of course, those conversations have always been one-way. 

But that’s not the case for all Super Brain services. The company also offers deepfaked video calls in which a company employee or a contract therapist pretends to be the relative who passed away. Using DeepFace, an open-source tool that analyzes facial features, the deceased person’s face is reconstructed in 3D and swapped in for the live person’s face with a real-time filter. 

SUPER BRAIN

At the other end of the call is usually an elderly family member who may not know that the relative has died—and whose family has arranged the conversation as a ruse. 

Jonathan Yang, a Nanjing resident who works in the tech industry, paid for this service in September 2023. His uncle died in a construction accident, but the family hesitated to tell Yang’s grandmother, who is 93 and in poor health. They worried that she wouldn’t survive the devastating news.

So Yang paid $1,350 to commission three deepfaked calls of his dead uncle. He gave Super Brain a handful of photos and videos of his uncle to train the model. Then, on three Chinese holidays, a Super Brain employee video-called Yang’s grandmother and told her, as his uncle, that he was busy working in a faraway city and wouldn’t be able to come back home, even during the Chinese New Year. 

“The effect has met my expectations. My grandma didn’t suspect anything,” Yang says. His family did have mixed opinions about the idea, because some relatives thought maybe she would have wanted to see her son’s body before it was cremated. Still, the whole family got on board in the end, believing the ruse would be best for her health. After all, it’s pretty common for Chinese families to tell “necessary” lies to avoid overwhelming seniors, as depicted in the movie The Farewell. 

To Yang, a close follower of the AI industry trends, creating replicas of the dead is one of the best applications of the technology. “It best represents the warmth [of AI],” he says. His grandmother’s health has improved, and there may come a day when they finally tell her the truth. By that time, Yang says, he may purchase a digital avatar of his uncle for his grandma to talk to whenever she misses him.

Is AI really good for grief? 

Even as AI cloning technology improves, there are some significant barriers preventing more people from using it to speak with their dead relatives in China. 

On the tech side, there are limitations to what AI models can generate. Most LLMs can handle dominant languages like Mandarin and Cantonese, but they aren’t able to replicate the many niche dialects in China. It’s also challenging—and therefore costly—to replicate body movements and complex facial expressions in 3D models. 

Then there’s the issue of training data. Unlike cloning someone who’s still alive, which often involves asking the person to record body movements or say certain things, posthumous AI replications must rely on whatever videos or photos are already available. And many clients don’t have high-quality data, or enough of it, for the end result to be satisfactory. 

Complicating these technical challenges are myriad ethical questions. Notably, how can someone who is already dead consent to being digitally replicated? For now, companies like Super Brain and Silicon Intelligence rely on the permission of direct family members. But what if family members disagree? And if a digital avatar generates inappropriate answers, who is responsible?

Similar technology caused controversy earlier this year. A company in Ningbo reportedly used AI tools to create videos of deceased celebrities and posted them on social media to speak to their fans. The videos were generated using public data, but without seeking any approval or permission. The result was intense criticism from the celebrities’ families and fans, and the videos were eventually taken down. 

“It’s a new domain that only came about after the popularization of AI: the rights to digital eternity,” says Shen, the Tsinghua professor, who also runs a lab that creates digital replicas of people who have passed away. He believes it should be prohibited to use deepfake technology to replicate living people without their permission. For people who have passed away, all of their immediate living family members must agree beforehand, he says. 

There could be negative effects on clients’ mental health, too. While some people, like Sun, find their conversations with avatars to be therapeutic, not everyone thinks it’s a healthy way to grieve. “The controversy lies in the fact that if we replicate our family members because we miss them, we may constantly stay in the state of mourning and can’t withdraw from it to accept that they have truly passed away,” says Shen. A widowed person who’s in constant conversation with the digital version of their partner might be held back from seeking a new relationship, for instance. 

“When someone passes away, should we replace our real emotions with fictional ones and linger in that emotional state?” Shen asks. Psychologists and philosophers who talked to MIT Technology Review about the impact of grief tech have warned about the danger of doing so. 

Sun Kai, at least, has found the digital avatar of his mom to be a comfort. She’s like a 24/7 confidante on his phone. Even though it’s possible to remake his mother’s avatar with the latest technology, he hasn’t yet done that. “I’m so used to what she looks like and sounds like now,” he says. As years have gone by, the boundary between her avatar and his memory of her has begun to blur. “Sometimes I couldn’t even tell which one is the real her,” he says.

And Sun is still okay with doing most of the talking. “When I’m confiding in her, I’m merely letting off steam. Sometimes you already know the answer to your question, but you still need to say it out loud,” he says. “My conversations with my mom have always been like this throughout the years.” 

But now, unlike before, he gets to talk to her whenever he wants to.

China’s next cultural export could be TikTok-style short soap operas

Until last year, Ty Coker, a 28-year-old voice actor who lives in Missouri, mostly voiced video games and animations. But in December, they got a casting call for their first shot at live-action content: a Chinese series called Adored by the CEO, which was being remade for an American audience. Coker was hired to dub one of the main characters.

But you won’t find Adored by the CEO on TV or Netflix. Instead, it’s on FlexTV, a Chinese app filled with short dramas like this one. The shows on FlexTV are shot for phone screens, cut into about 90 two-minute episodes, and optimized for today’s extremely short attention span. Coker calls it “soap operas for the TikTok age.”

In the past few years, these short dramas have become hugely popular in China. They often span nearly a hundred episodes, but since each episode is only one or two minutes long, the whole series is no longer than a traditional movie. The most successful domestic productions make tens of millions of dollars in a few days. The entire market of short dramas in China was worth over $5 billion in 2023.

This success has motivated a few companies to try replicating the business model outside China. Not only is FlexTV translating and dubbing shows already released in China, but it has also started filming shows in the US for a more authentically American viewing experience.

It’s easy to compare apps like these to Quibi, a high-profile video service that infamously failed after less than a year in 2020. 

But these latest Chinese apps are different. They don’t aim for slick, expensive productions. Instead, they choose simple scripts, shoot an entire series in two weeks, market it heavily online, and move on to the next project if it doesn’t stick. 

“The biggest difference between short dramas and films is that they provide different things. We have to analyze the psychological needs of our audience and understand what they want to see … and we try to provide some emotional values,” Xiangchen Gao, the chief operations officer of FlexTV, tells MIT Technology Review. 

When a show finds the right audience, it can generate significant revenue in the US too. The top-grossing show on FlexTV can bring in $2 million a week, while the production costs less than $150,000, Wang says.

Several other apps, like ReelShort and DramaBox, are also racing to bring Chinese short dramas to an international audience. They frequently top app stores’ download charts and produce blockbuster shows. Short dramas have been proven to work in China. It’s not always easy to replicate a business model in a different market, but if they succeed, they could be China’s next big cultural export.

The roots in Chinese web novels

Short dramas like Adored by the CEO are often adapted from another cultural product that is distinctly Chinese: web novels.

Web novels are a unique form of literature that has been popular on the Chinese internet for much of the last two decades: long stories that are written and posted chapter by chapter every day. Each chapter can be read in less than 10 minutes, but installments will keep being added for months if not years. Readers become avid fans, waiting for the new chapter to come out every day and paying a few cents to access it.

While some talented Chinese book authors got their big break by writing web novels, the majority of these works are the popcorn of literature, offering daily bite-size dopamine hits. For a while in the 2010s, some found an audience overseas too, with Chinese companies setting up websites to translate web novels into English.

But in the age of TikTok, long text posts have become less popular online, and the web-novel industry is looking to pivot. Business executives have realized they can adapt these novels into super-short dramas. Both forms aim for the same market: people who want something quick to kill time in their commute, or during breaks and lunch.

Many of the leading Chinese short-drama apps today work closely with Chinese web-novel companies. ReelShort is partially owned by COL Group, one of the largest digital publishers in China, with a treasure trove of novels that are ready for adaptation.

COURTESY OF FLEXTV

To get a quick sense of what these stories are like, you just need to take a look at their titles: President’s Sexy Wife, The Bride of the Wolf King, Boss Behind the Scenes Is My Husband, or The New Rich Family Grudge.

One of the highest-grossing shows on FlexTV is called Mr. Williams! Madame Is Dying. It’s a corny romance story about a love triangle, ultra-rich families, cancer, rebirth, and redemption, and it was adapted from a Chinese web novel that has nearly 1,300 chapters. The original story has been turned into a Chinese short drama, but FlexTV decided to shoot another version in Los Angeles for an international audience.

These short dramas prioritize quick, oversimplified stories of love, wealth, betrayal, and revenge, sometimes featuring mythical creatures like vampires and werewolves. Stories of marrying into a rich family attract men, while stories with a powerful female protagonist in control of her life appeal to women, says Gao, the COO of FlexTV. 

“Quibi mostly served the [artistic] pursuits of directors and producers. They thought their tastes were better than the general public and their work was to be appreciated by the elites,” he says, “What we are making is more like fast-moving consumer goods. It’s rooted in the needs of ordinary users.”

Translating the story to a US audience

Still, no matter how universal the plots are, these short dramas need to be adapted to their local audience. 

Ty Coker’s work is one example. The character they voiced, a personal assistant to the male protagonist, was named Dawei Hu in the original Chinese production. But in the English dubbing, “Dawei Hu” became “David Hughes.” All the other characters, as well as the geographical references, received similar treatments, while the visuals didn’t change. Sometimes the results of this straightforward swap method are a little jarring. “They would mention: ‘Oh, so and so is coming in from New York. And it was like, okay, I don’t think that character is originally from New York, but we’ll roll with it,” Coker says.

This “Americanization” could make it easier for some viewers to follow and remember the show. “My mom, who watches a lot of soap operas—she’d probably find it easier to follow if everyone has American names,” Coker says. It reminds Coker of anime shows they watched as a child, which would substitute Western names for the Japanese ones.

But that’s not where the localization efforts end. With shows that are later dubbed into a different language, there’s always going to be a feeling of mismatch. That’s why short-drama platforms are now filming their own productions with translated scripts and non-Chinese actors, sometimes even in Hollywood. It costs much more than dubbing an already-made show, but they believe it’s worth it.

FlexTV has just finished filming a new show in Los Angeles. Named Lost in Darkness, it is about a visually impaired woman trying to figure out which of two male characters murdered her father. The production took 10 days in total, with 34 actors filming over 150 scenes. It cost between $150,000 and $200,000. 

COURTESY OF FLEXTV

Roger Chen, a producer of the show, has been instrumental in bringing short dramas to the US. He originally produced traditional-length movies and animations in the States but formed a new company called Purple Filter last year, after noticing the demand for Chinese-adapted stories. His company now coordinates between Chinese platforms and the LA filming industry, scouting actors, directors, and producers who are interested in this new form of content.

He admits these shows can be a hard sell for actors in the US. 

“An episode of a short drama is only one and a half minutes long, so the plot buildup is short, if there is any … Actors who are accustomed to traditional scripts find it hard to accept,” Chen says. “But we discovered that it’s necessary to have strong conflicts and quick plot twists. This is the unique content that suits the smartphone medium.”

His company has worked with most of the major platforms, including DramaBox and FlexTV. Meanwhile, there are more and more teams like Chen’s joining the American short-drama industry. “We believe that media content for phones will become a mainstream trend,” he says.

A well-oiled business in targeted ads 

The business model of these short dramas is one that American TV audiences are mostly not familiar with. Unlike most American streaming services, which require subscriptions, Chinese platforms for streaming and web novels use a business model of paying by the episode or chapter. Essentially, the first 10 or so episodes are always available for free, but once users are hooked, they need to pay a certain amount to watch each episode. It resembles the microtransaction mechanism in mobile games, which Chinese companies, like the developer of the global hit Genshin Impact, also perfected. Users can quickly rack up thousands in payments by buying small items in-game here and there.

FlexTV has a similar tactic. You can pay $5 for 500 in-app coins, which in return unlock about seven episodes. A whole series therefore can cost around $50, but there are also small tasks users can do in the app to earn free rewards, like watching ads, posting about the app on social media, and doing daily check-ins. 

Evidently, some people are willing to pay, as some of these shows are starting to see serious revenues. Gao shares that Mr. Williams! Madame Is Dying is bringing in over $2 million in user payments a week. Similar apps have also reported significant revenues. ReelShort made $22 million in December 2023, the company told the Wall Street Journal, and its global download count has already surpassed that of the ill-fated Quibi. 

But there’s a caveat: these phenomenal numbers are also driven by heavy ad spending, which is an essential part of the business model. While it’s cheap to make these shows (around $150,000 per show if filmed in the US), millions of dollars are then spent on pushing them to prospective audiences. For Mr. Williams! Madame Is Dying, for example, the company had to spend $1 million on ads to get over $2 million in revenues. 

FlexTV relies on targeted advertising through major platforms like Google, Meta, and TikTok. These platforms allow them to choose what kind of audience they want the ads to impress. For Mr. Williams! Madame Is Dying, FlexTV targeted “young women between the ages of 20 and 40, who like romance content and reading,” says Gao.

For each dollar spent on advertising, the target return on investment is at least $1.30 to $1.50, says Gao. The viral popularity of short dramas in China is no accident but the result of a well-oiled digital marketing industry that has adjusted to the TikTok age. 

Despite early successes, the short-drama business inside China is now starting to run into difficulties. As its popularity grows, the Chinese government has started to censor it much as it does the film and TV industry. In a three-month period ending in February 2023, the government banned 25,300 short-drama series (totaling 1.3 million episodes) for being too violent, too sexually suggestive, or too trashy. Chinese short-video platforms like Douyin and Kuaishou have since been routinely removing shows or restricting their producers from buying targeted ads. 

This is also part of the reason why companies like ReelShort and FlexTV are looking to expand in foreign markets, where the risks of censorship are much lower. The short-drama industry in the US is akin to what it was like in China in 2021, says Gao, which means there’s little competition and plenty of room to grow. The company plans to produce shows in six languages in the future, but North America will remain its most important market. 

Ultimately, they’re betting that these dopamine-inducing mini soap operas will appeal to audiences regardless of their culture or language. “We don’t think Chinese and American audiences have fundamental differences in taste,” Gao says. “Chinese web novels have studied human nature deeply, and they do a great job at evoking emotions. Scriptwriters and directors who excelled in China can capture everyone’s universal desires.”

Meet the divers trying to figure out how deep humans can go

Two hundred thirty meters into one of the deepest underwater caves on Earth, Richard “Harry” Harris knew that not far ahead of him was a 15-meter drop leading to a place no human being had seen before. 

Getting there had taken two helicopters, three weeks of test dives, two tons of equipment, and hard work to overcome an unexpected number of technical problems. But in the moment, Harris was hypnotized by what was before him: the vast, black, gaping unknown. 

Staring into it, he felt the familiar pull—maybe he could go just a little farther. Instead, he looked to his diving partner, Craig Challen, floating a few feet to his right. They had both been diving increasingly dangerous and unplumbed caves for years, making them two of only a handful of people with the skills to assist in the rescue of the Thai soccer team that got trapped in one in 2018. They knew extreme risk, and each other, well. Even through the goggles and the mouthpiece of the breathing apparatus, its four thick hoses curling around his face like mammoth tusks, Harris could see that Challen felt the same way. They both wanted very badly to push forward into the dark expanse.

Instead, on Harris’s cue, they turned back. They weren’t there to exceed 245 meters—a depth they’d reached three years earlier. Nor were they there to set a depth record—that would mean going past 308 meters. They were there to test what they saw as a possible key to unlocking depths beyond even 310 meters: breathing hydrogen.

The problem has existed for more than a century: How can a human body withstand underwater pressure significantly past its natural threshold? Naval units and offshore oil companies around the world have long been invested in figuring it out for power and profit, and in the 1970s and ’80s their research began filtering into the civilian world, where people were testing the limits of their own curiosity.

COURTESY OF RICHARD HARRIS

This included people like Sheck Exley, a high school math teacher in Live Oak, Florida. Exley became an international icon in the diving community for his record-breaking dives—exceeding, in some cases, the limits of the military and commercial professionals. He had been diving the underwater caves of North Florida since he was a teen—by 1972, at age 23, he was the first person in the world to log 1,000 cave dives—and exploring them is what pushed him to go deeper. He’d been hooked since his first cave dive in Crystal River, Florida, in 1966. “I kind of wandered off into the cavern there, my eyes adjusted, and I swam a little bit further, peering off into the darkness,” he told AquaCorps, a magazine for divers, in 1992. “I guess I’ve been peering off into that darkness ever since.”

At around 40 meters, breathing the gas mixture we call air—78% nitrogen, 21% oxygen, and 1% trace gases—causes inert gas narcosis, or the “martini effect,” named for the incapacitating state it induces. A little deeper and oxygen becomes toxic. For years, the US and British navies had been using helium to dilute the oxygen and nitrogen in a diver’s tank as a means of counteracting both these problems, but few outsiders knew about it. In 1981, after the German cave diver Jochen Hasenmayer reached 143 meters using a helium mixture, Exley started using it as well—despite the knowledge that only a few years earlier, two divers in Florida had experimented with the mixture and died. 

Divers who go past 40 meters typically don’t use a constant gas ratio—they cycle through mixes of nitrogen, oxygen, and helium as they descend and ascend, modifying them according to location, water temperature, neurological tolerance to narcosis, and many other variables. Decompression tables, which lay out different gas mixtures and the amount of time to be spent breathing them, provide a precise road map for this process—very necessary, as ascending too quickly releases the accumulated gases in a diver’s blood and tissue the way unscrewing the cap on a bottle of soda releases bubbles, resulting in the painful, often debilitating condition known as the bends. Without the tables, ascending from great depths was too risky. But generating them required high-level technology and archives of data unavailable to most people.

Exley was able to persuade a friend who worked in commercial diving to give him a sample table, which he used to extrapolate past 121 meters, working everything out on his computer. In 1987, guided by this information, he used helium to successfully break the 200-meter diving barrier at Nacimiento Mante in Mexico—a 24-minute descent that forced him to stay underwater for 11.5 hours to decompress. Over that period, he felt himself becoming dangerously weak. His blood sugar dropped, the cold seeped into his limbs, and his exposed hands and face began to grow wrinkly, become raw, and then flake. “At the time I felt I could have dived deeper,” he told InDepth magazine that year. “But I knew that I had reached my decompression limits.” 

In 1988, Exley beat his own record, hitting 237 meters using a more precise table generated by the physiologist Bill Hamilton. Known among divers as “The Prince of Gases,” Hamilton had been involved with some of the earliest work on decompression past 200 meters. 

He had helped develop a computer program that generated extremely precise tables calibrated to any number of parameters, which he created for commercial diving companies and navies around the world. “Through them, he had staggering amounts of data,” says Bill Stone, an aerospace engineer and cave diver who built and sold an early rebreather—a type of breathing apparatus that recycles exhaled air by scrubbing it of carbon dioxide. “He processed all that and developed stochastic models that figured out where you could gain speed on decompression and not hurt yourself.” In the ’80s, Hamilton made the unprecedented move of creating custom tables for people who were not commercial or military divers, forever endearing him to the diving community. 

But in 1994, Exley died attempting to reach the bottom of the Zacatón sinkhole in Mexico, 332 meters down—he got to 270 meters. A likely factor was high-pressure nervous syndrome (HPNS), a neurological condition involving uncontrollable tremors brought about when a diver descends too rapidly at high pressure past 150 meters. Adding nitrogen to the mix of helium and oxygen could help—nitrogen’s narcotic effects could dampen the symptoms. But at that depth, adding more nitrogen would make it more difficult to breathe because of the gas’s density, and the narcosis would be debilitating. 

Divers still pushed past Exley’s record into 250 and then 300 meters, white-knuckling through the HPNS symptoms. But at such depths, helium becomes too heavy for the human body to process. “As you go deeper, for every 10 meters you get an extra atmosphere of pressure—so once we’re down at those depths of 250 meters, that’s 26 atmospheres,” Challen, Harris’s dive partner, told me. “It becomes a physical problem of moving the gas in and out of your lungs.” In order to alleviate this problem, divers would need to breathe something lighter than helium. 

“And we’ve only got one lighter gas to use,” Challen says. “After hydrogen, there’s nothing.”

What became known as the H2 Working Group began as a “thought experiment,” according to Michael Menduno, a veteran cave diver and editor in chief of InDepth magazine, who brought everyone together in collaboration with John Clarke, the former scientific director of the US Navy Experimental Dive Unit. It was also a way to socialize the way scientists, engineers, and tinkerers know how: by arguing over how to make something work. 

The idea for the working group emerged in 2020, when Menduno, while researching an article on deep dives past 250 meters, concluded that the diving world had reached its limit with helium. Exley himself had pointed to what he saw as the next step in reaching unprecedented depths: “From what I’ve been learning, there seems to be some real potential for hydreliox”—that is, an oxygen-helium-hydrogen mix. Still, “no one has looked at its use for deep bounce dives,” he told Menduno in a 1992 interview, referring to dives without the use of a chamber to help the diver reach a safe gas saturation point before venturing into the water.

When the group formed in that year, at the beginning of the pandemic, Menduno was familiar with previous attempts to use hydrogen. In the ’90s he had traveled to France to watch a team of medical doctors, engineers, scientists, and trial divers from Compagnie Maritime d’Expertises (Comex), a Marseilles-based company focusing on deep diving systems and equipment, attempt to pass 701 meters in a hyperbaric chamber pumped with a hydrogen gas mixture. (It required 15 days of compression and 23 days of decompression.) This was Hydra 10, part of a series of experiments using hydrogen. After Hydra 12, in 1996, the project had scrapped pursuing hydrogen altogether. Companies were turning to manned underwater vehicles and saturation diving—that is, staying at extreme depths long enough to allow the inert gas to fully dissolve in a diver’s blood, which is only possible with a chamber. “[In commercial diving] there’s no poetry, no sense of humor—you just want to be efficient,” says Jean-Pierre Imbert, who worked on the Hydra project and had invited Menduno to observe the experiments in the ’90s.

In 2012, building on Comex’s work, a team of Swedish divers, called the Hydrox Project, had successfully breathed a mixture of hydrogen and oxygen for five minutes at 40 meters. They were hesitant to share the information, concerned that people without the expertise might make the attempt at their peril. “We didn’t publish or tell too much about how we did it, because we believe if you do this wrong—you have to understand it, or you will fail miserably,” diving engineer Åke Larsson, one of the members of the Hydrox Project, later said. “But it’s not rocket science. If you do your homework, it’s not complicated. But you have to do it right.” 

That was the history on Menduno and Clarke’s minds when they began calling up experts in the field to discuss whether it would be possible for a diver outside a chamber to descend beyond 200 meters breathing hydrogen. Each member was an expert in a different element of what would be involved in such a dive—the mechanics of the breathing apparatus, the calculation of gas mixtures, the physiological issues to be overcome as divers coped with cold and pressure. By the end of the first meeting, they had come up with so many potential problems they created what they called a “challenge tree” to organize them; members gravitated toward their areas of expertise and came back to present possible solutions to the group. The tree kept branching; solutions begat new problems. 

SIMON MITCHELL

Harris joined the working group at Menduno’s invitation. The team of divers he was a part of, the Wetmules, had reached 245 meters at the Pearse Resurgence, the source of the Pearse River in New Zealand, but the cave stretched on, and they wanted to map it. They knew that helium was not going to get them past 300. Hydrogen, however, was nerve-racking. “I don’t really want to be the guinea pig who famously popped under water after switching to hydrogen,” he told the group at the first meeting. “So it’d be nice if we can solve that.”

Hydrogen—as the Hindenburg disaster showed the world—is highly flammable. A spark hundreds of times smaller than the slightest amount of static you can feel on your fingertip would be enough to ignite it. And even if such a spark didn’t cause an explosion, it could still set the gas on fire. “If you’re breathing that mix when it’s burning,” Clarke told the group, “it’s going to be a very unpleasant dive.”

David Doolette, a research physiologist at the US Navy Experimental Diving Unit who had himself dived the Pearse Resurgence in the ’90s, was skeptical but considered himself “dihydrogen-curious,” referring to the molecular structure of hydrogen gas. For one thing, its thermal properties mean divers risk hypothermia. “It’s gonna suck the heat out of your body to a dangerous level,” he told the group. And there was no data available to calculate the appropriate decompression times. “It’s going to be a challenge to do the decompression calculations,” he told the group. “Well, not a challenge doing that—that’d be easy. It’d be a challenge to get them right so that they were safe to dive.” 

SIMON MITCHELL

Others shared his skepticism. Nuno Gomes, a former world-record-holding cave diver, was particularly concerned about decompression. “I think if we’re going to go with this, we’re going to have to progress slowly,” he told the group. “Start with shallow dives and progress to deeper and deeper dives.” But according to Doolette and Stone, the aerospace engineer, it soon became clear that Harris had already decided to try hydrogen in the Pearse Resurgence. At a conference in Australia in 2022, as members of the H2 Working Group met up to talk in person, Imbert said that going beyond 3.5% hydrogen would likely trigger a detonation. 

“Harry nods and goes, ‘Well, I don’t think that’s true,’” Stone told me. Imbert asked how he could prove that. “Harry says, ‘Well, last week I did 7% in my pool.’ Everybody perked up.” 

Harris had ordered a canister of hydrogen delivered to his suburban home in Adelaide and, as he later explained, “decided to have a bit of a play with it.” He rigged his rebreather for hydrogen and put it in his backyard pool, hoping to contain any potential blast. He filled the rebreather with hydrogen and then, backing way from the pool, began to introduce oxygen. (His dog observed from outside the pool fence; his wife was out.) 

When nothing exploded, he started adjusting the ratio of oxygen and hydrogen, becoming confident enough to try using the rebreather himself. His first sip, he later told me, felt light, slippery, and cold. It was almost delightfully easy to breathe. “Hydrogen voice is much sillier than helium voice,” he told me. “And I was pleased the house and the dog were intact.”

The others were amazed. Some were perturbed. “Everybody has to make this decision for themself,” Stone told me. “The Pearse Resurgence is not a place to experiment. When you go in there, you should be using gear and techniques that you know are going to work at that depth. You don’t want to be doing physiological experiments at 300 meters’ depth. That’s what killed all of the other divers who went beyond 200 meters’ depth. So my advice to Harry and anybody else who wants to play this game is the same as what I gave Exley: Go. To. A chamber. Simulate this first.”

“The group was sort of split,” Menduno told me. “I mean, everybody was supportive of Harry, but there were some people in the group that thought: You’re gonna die. Some of the people in the group were upset and worried that their friend was going to go off and do this thing and potentially die.”

Around the first corner of the Pearse Resurgence, the light disappears, as though the dark walls, black marble striated with veins of gray quartz, have absorbed it. The cave sometimes narrows so much that if you stood, you could touch the ceiling. Other parts billow out into enormous chambers. At one point, jagged fingers of rock bristle from the walls. Other, deeper parts of the cave are smooth and almost perfectly round, broken only by dark fissures that lead to unexplored tunnels. 

As each section of the cave gets discovered, it receives a name. Going down in February 2023, Harris and Challen passed through the Nightmare Crescent, the Needlebender, the Gargleblaster, Weaver’s Ledge, the Big Room, and finally the Brooklyn Exit. The water was 6 °C and perfectly clear. Aside from the brief hisses and clicks of the rebreathers—the crackle of the solenoid triggering, the sigh of gases being pumped through the loop—there was an otherworldly silence.

At 120 meters, the cave opens up onto a plateau that drops off into an abyss. “At that point it’s like standing on the precipice,” Harris told me. “And it feels like you are really beginning the journey.” 

The abyss takes you down 50 meters through a vertical tunnel. By 170 meters, Harris could track where he was on the map in his head, following familiar rock formations. They wanted to preserve their energy and prevent carbon dioxide buildup in their joints, so they limited their movement, relying on underwater scooters to move. They slowly tied off at different points in the descent, working around ropes left behind from dives past, some of which had been installed by Doolette 20 years before. 

At 230 meters, Harris had done something nobody had done before—swimming freely to unimaginable depths and breathing in hydrogen.

Harris remembers that even though his mind was absorbed with their strict plan, hypervigilant to any strange noises from his rebreather that could mean failure, he took a moment to pause, thinking: “What if I never got to see this again?” 

At 200 meters, Harris introduced the hydrogen. For the next 30 meters he gauged his body’s reaction. He was calm, clearheaded, but even more, he noticed that the light tremors in his hands he usually got at this depth, an early sign of high-pressure nervous syndrome, had disappeared. He looked to Challen, who was using helium, as he tied off the rope: his dive partner’s hands had a visible tremor. 

At 230 meters, Harris had done something nobody had done before—swimming freely down to these unimaginable depths and breathing in hydrogen—but his eyes were trained just a little farther, to the unexplored drop only 15 meters away. “I would be lying if I didn’t say I dream about going down there,” he later told an audience in Malta. 

Those obsessed with learning how to put the human body under incredible amounts of pressure make up a relatively small community; for the most part, everybody is one degree of separation from everyone else. Clarke, the Navy scientist, and Susan Kayar, a physiologist and expert in decompression who was also part of the H2 Working Group, had both worked at the Navy Medical Research Institute in the ’90s. They had lost touch, but they reconnected after both published novels that included a scene with a hydrogen dive—hers for a rescue mission, his to recover a UFO that had crashed into the sea. “Hers is more realistic in that regard,” Clarke says. “I hadn’t spoken to Susan in 30 years, and then we finally found out: Gee, we’re thinking very much alike!”

SIMON MITCHELL

Doolette, the research physiologist skeptical about using hydrogen, has known Harris for more than 20 years, ever since Harris took Doolette’s diving medicine qualification course in Australia. At the time, they were also both working in the hyperbaric medical unit at the Royal Adelaide Hospital. Harris wanted to get more advanced at cave diving and Doolette was a longtime technical cave diver.

Even before the H2 Working Group began meeting, Doolette knew that Harris was thinking about diving with hydrogen. “I knew things were being planned,” Doolette told me. “It probably came as a surprise to some of the workshop, but not to me … I knew it wasn’t theoretical.”

But Doolette felt that testing hydrogen on a deep cave dive was too risky. “There’s a whole hydrogen industry—and their approach is to exclude hydrogen and oxygen from getting near each other. So when you’ve got to mix them together—well, there aren’t even procedures, really,” he says. “I certainly thought it was foolhardy.” 

Stone, too, was skeptical of the expedition—for any dive beyond 200 meters, he said, the concept should be tested first. “I’ve been doing this for a long time, and I’ve got a lot of dead friends,” he says. 

“We’re moving into the zone where robotics isn’t science fiction,” he continues. “It’s hardware, it’s software—we’re doing it. Is it perfect? No. But it’s a couple years away. We’re close to doing dives longer than the longest human dives, and with a kilometer-depth rating it’ll be far, far beyond even hydrogen diving.”

As a diver, Doolette says, he understands wanting to go where nobody else had been before: “Cave divers and other people explore for the joy of it. A robot’s not gonna get that.” The other draw, he says, is the puzzle of figuring out where the cave is going and how to follow it. And when you finally get into the water, the rest of the world falls away. 

A few weeks after the hydrogen dive, Harris gave a presentation to the H2 Working Group, apologizing for “sneaking around a little bit.” He concluded with a caveat: n = 1. “Meaning it has been successful one time,” he said. When the PowerPoint ended, the group applauded. “Just something you can add to your final slide, Harry,” Doolette said. “The probability of survival is greater than zero.” 

Doolette was relieved when he learned that Harris had made it back to the surface. “I recognized that it was a pretty groundbreaking, momentous dive,” he told me. “I don’t see the purpose of going deep for the sake of going deep, but if you’re committed to an exploration project and it involves cave diving in pursuit of exploration and discovery, then that’s what you do.”

“There’s always a little temptation to go that little bit further. That’s why we’re doing this stuff,” Challen told me. “We have this defective aspect to our characters that urges us to push on just a little bit more.” 

Samantha Schuyler is a writer, editor, and fact-checker living in New York.