How China Has Revolutionized AI

A world completely run by robots and artificial intelligence? Such an idea sounds like something out of a far-fetched sci-fi story. Yet, it has come even closer to reality than we might like to think. China has already begun pushing artificial intelligence (AI) development as one of the cornerstones of its vision for a Chinese future. In fact, the Chinese Communist Party has set a goal to make significant contributions to AI research by 2020 and plans to lead the world in AI innovation and technology by 2030. This technology can already be seen at many levels of Chinese society, including in some select courtrooms.

Earlier this year, the Beijing Internet Court introduced an online litigation center featuring an AI judge and a mobile court, both of which allow court hearings to be handled online. The AI takes on the appearance of a female judge, and its voice and facial expressions match those of a real person, granting it a level of legitimacy unmatched by purely artificial constructs.

However, it seems like the AI judge won’t be making any verdicts just yet. According to an official statement made by the Beijing Internet Court, the AI judge will only be used for the completion of “repetitive basic work,” which includes litigation reception and online guidance. While there’s no doubt that China has bigger and bolder plans for its AI industry, it seems this particular facet of it has been put on hold.

But AI is increasingly interwoven with education as well. China’s investments in artificial intelligence-enabled teaching have recently skyrocketed in scale and funds, and millions of students use some form of AI to learn core subjects. In a learning center opened by Squirrel, one of the largest AI education companies in China, students and teachers alike rely on laptops for instruction. Students work on practice problems and submit their answers online, and teachers monitor their progress through a real-time dashboard.

Squirrel offers help on annual standardized tests, and more than 80 percent of its students return each year. The system is designed to collect as much data on its students’ progress as possible, allowing for lesson plans to be tailored to each student’s needs. A student begins by taking a diagnostic test, which assesses the student’s understanding in various key concepts. As the student studies, the AI-assisted system will update its curriculum accordingly. The company has opened 2,000 learning centers in the past five years, and has raised over $180 million in funding from investors.

Though Squirrel is not the first company to create an AI tutor, it has been one of the more successful companies to do so. Squirrel’s team works with teachers to divide each subject into small conceptual pieces to help find gaps in each student’s understanding of the material.

Squirrel has yielded impressive results in improving test scores. In a four-day study conducted in October 2017, it was found that the system was better at improving the math test scores of 78 middle school students than teachers working in a traditional classroom. Squirrel’s founder, Derek Li, hopes to expand his tutoring service by implementing his curriculum directly into classrooms across the country. The company has already arranged with several schools in China to make Squirrel their main method of instruction, and Li plans to expand its operations greatly in the near future.

AI is set to become integral parts of both the courtroom and the classroom, but its utility does not end there. The technology has even been used to solve, of all things, a murder. The suspect, who has only been identified as Mr. Zhang, was accused of strangling his girlfriend with a rope in Xiamen on April 11, and was also accused of identity theft after pretending to be his girlfriend and contacting her employers through her WeChat account.

When Zhang arrived in his hometown the next day, he tried to apply for a loan using his girlfriend’s face through an app called Money Station, which uses AI to verify the applicants’ identities. But the facial recognition technology found no signs of eye movement, and staff at the lender contacted police after the camera recognized bruises and red marks on the woman’s face.

Though AI has certainly improved many aspects of life in China, concerns have been raised about how the government—and the futuristic corporations it backs—will use the massive amounts of data about Chinese citizens collected each day. Critics warn that the growth of AI applications could infringe upon civil liberties, however limited those are in China. “I’m deeply concerned that AI is not going to empower the people, but instead that the government will use AI to further suppress its citizens, especially in combination with surveillance, big data, and machine learning,” said Lokman Tsui, an activist at the Chinese University of Hong Kong. AI can be a double-edged sword; while it makes processes in many fields more seamless and convenient, it can be done at the expense of privacy and civil liberties.

China has made great strides in AI research, and its simultaneous advances in health care, transport, and communications seem promising. If AI development in China continues to progress at its current rate, however, all of that could come at a cost.