A 75-year-old Harvard grad is propelling China’s AI ambitions

One of many discipline’s most influential figures is Andrew Chi-Chih Yao, whose training {and professional} life have straddled the world’s two greatest economies. China-born and Harvard-trained, Yao is his nation’s solely recipient of the Turing Award, pc science’s equal of a Nobel Prize. After virtually 40 years within the US, he returned to China in 2004. Now he teaches a prestigious but little-known college class that has formed a few of the nation’s greatest AI startups, knowledgeable authorities coverage and molded a technology of teachers.

“We’ve got an excellent alternative within the subsequent 10 or 20 years, when synthetic intelligence will change the world,” Yao stated in Might 2019. He urged China to “take a step forward of others, to domesticate our abilities and work on our analysis.” The scientist, who not often speaks to overseas media, didn’t reply to Bloomberg’s requests for an interview.

The “Yao Class”—an undergraduate pc science course at Beijing’s Tsinghua College, alma mater to President Xi Jinping and lots of of China’s ruling elite—has exerted a profound influence on the nation’s expertise pioneers and rising scientific prowess. Its graduates kind a robust community throughout the nation, advising on every others’ initiatives and pooling assets and capital the place wanted.

Yao’s acolytes have created startups price greater than $12 billion at their peak, together with Alibaba-backed facial-recognition large Megvii Know-how Ltd. and Guangzhou-based Pony.ai Inc. Others train at top-flight American universities together with Stanford and Princeton.

“Simply his willingness to come back again to China means quite a bit,” stated Hu Yuanming, a Yao class pupil from 2013 to 2017 and the chief govt officer of pc graphics startup Taichi Graphics Know-how Inc. His firm is backed by Sequoia China, Supply Code Capital, GGV Capital and BAI Capital, having completed its collection A spherical of financing of $50 million in February.

Hu is a beneficiary of the Yao Class expertise pool. Pony.ai’s Lou Tiancheng and Megvii’s Tang Wenbin suggested Hu on founding his firm, and he says hiring is simpler for Taichi than for a lot of small companies. Present Yao college students have undertaken internships too.

Driving Drive

One factor the US and China agree on is the huge potential of AI—a sweeping discipline that can outline a lot future expertise and which is now a key battleground in Washington and Beijing’s battle for tech ascendancy. Given its potential for making weapons smarter, AI might also have main nationwide safety implications.

The US Nationwide Safety Fee on Synthetic Intelligence, chaired by former Google CEO Eric Schmidt, warned final 12 months of the dangers inherent in China’s rising grasp of the sphere. “If the USA doesn’t act, it should doubtless lose its management place in AI to China within the subsequent decade and turn into extra susceptible to a spectrum of AI-enabled threats,” the NSCAI report stated.

In the meantime China has framed AI as a “core driving power” in its industrial transformation and a “new focus of worldwide competitors” because it pushes for technological self-reliance. In 2017, the nation set a goal for AI-related industries to achieve 1 trillion yuan ($148.2 billion) by 2030.

With the world’s largest pool of Web customers and an unprecedented quantity of information, China has had marked—and controversial—success in AI, particularly in fields like facial recognition. Corporations like SenseTime Group Inc. and CloudWalk Know-how Co. are among the many sector’s most superior globally. China’s share of worldwide AI patent submitting reached 52% in 2021, up from 12% in 2010, in response to analysis from Stanford.

Some specialists say China’s AI experience is restricted in scope and extra centered on home surveillance than world domination. However no matter whether or not China involves dominate AI, or merely maintains its place as one of many prime gamers, Yao is a key a part of the nation’s toolkit.

[also_read title = “Also Read” article_title= “China launches second space station module, Wentian” id = “8048821” liveblog = “no”

Born in 1946, Yao emigrated to Taiwan as a child. He has described the upbringing he shared with his two siblings as happy and middle-class with an emphasis on traditional Chinese values, including education. He was an excellent student, and has said he considered scientists Galileo and Newton to be heroes and found physics more creative than Sherlock Holmes mysteries.

He moved to the US in 1967 to study Physics at Harvard University and credits his wife, Frances Yao, with introducing him to algorithms. A former PhD student at MIT, she is now also a professor of computer science at Tsinghua.

Yao taught for nearly three decades in the US, mostly at Stanford and Princeton, before returning to China in 2004. A number of other celebrated Chinese scholars returned from abroad around the same time, including Nobel prize-winning physicist Yang Chen-Ning and biophysicist Shi Yigong.

Yao told the state-run Xinhua News Agency that the opportunity to educate young Chinese students meant it was “not a difficult decision” to make.

New Understanding

Former students say Yao’s accessible and participatory teaching style helps to unlock the complex, highly abstract ideas at the heart of his discipline. He’s been known to invoke the Wizard of Oz or Alice in Wonderland when discussing his journey through computer science. Students are encouraged to answer questions on the spot and to challenge their teacher, and may be treated to KFC or Pizza Hut if a class member solves a particularly tough problem.

And the tough ones really are tough. Yao’s Millionaires’ Problem asks how two individuals can decide which of them is richer, if neither is prepared to say how much money they have. Answering such questions through cryptography—the study of secure communications techniques—has real-world applications for e-commerce, data mining, and many of the corners of the internet that call for passwords.

Within his field, Yao is perhaps best known for his work on the Min-Max Principle, a decision rule that is critical to game theory and computing.

“Professor Yao’s work gave us new ways of understanding algorithms,” said Aleks Kissinger, associate professor of quantum computing at the University of Oxford, in his introduction to a speech Yao gave in May. “The way that he explains fundamental problems is very relevant to scientists but also to anyone interested in more fundamental questions about the limits of what we can accomplish and what kinds of problems we can solve.”

In addition to his computer science class at Tsinghua, Yao has established more specialized classes in AI and quantum information. He also serves as the chief editor of China’s high-school AI textbook—a publication that was introduced in 2020.

“China missed the microelectronics revolution 70 or 80 years ago, so today it is difficult to catch up with the advanced level of the international semiconductor industry,” Yao said in an interview with China Global Television Network last year. “But in emerging fields such as quantum technology and artificial intelligence, China is expected to become an important player.”

Zou Hao, a former Yao student whose startup Tsimage Medical Technology offers AI-driven diagnosis services, believes Tsinghua graduates will play an important role in the technology’s future. “As time goes by, there will be more and more talents from Yao Class that make a difference and get great achievements,” he said.

Other entrepreneurs from the Yao Class include Li Chengtao, founder and CEO of AI drug discovery firmGalixir; Qi Zichao, co-founder and chief architect of metaverse startup DeepMirror and Long Fan, founder-president of blockchain startup Conflux. In academia, Yao alumni have been found on staff at Duke, Princeton and Stanford Universities, as well as at Tsinghua and Renmin University of China.

“Whether it’s applying for studying abroad or getting a job at universities, the label of Yao Class graduate did benefit me,” said Huang Zhiyi, an associate professor of computer science at the University of Hong Kong who was a Yao Class compatriot of Pony.ai’s Lou.

“Almost every aspect of my life and work is impacted by my experience there.”