Recently, China's state-run news agency, Xinhua, introduced a new broadcast news anchor. He is a product entirely...
The "English AI Anchor" is an animation based on the human anchor Zhang Zhao. It learns to read and speak news briefs fed to it using machine learning that's applied to hours of video of humans reading news. The news agency said the anchor will produce reports for its websites and social media profiles, while reducing costs.
It's yet another example of China's national strategy for AI at work. The country is aggressively investing in machine learning and other advanced technologies, pouring money into things like facial recognition, self-driving cars and AI chips, with the aim of making its economy more efficient and automated. The Chinese government's AI investment will soon outpace spending on AI by the U.S. and other nations, and it's raising questions about how countries will stay competitive in the 21st-century global economy.
"The [Chinese] government has an AI-first strategy," said Philipp Gerbert, senior partner and managing director at Boston Consulting Group. "China is really driving AI aggressively."
China powers ahead on AI investment
In a webinar hosted by the MIT Sloan Management Review, Gerbert cited numbers from a new report, "Global Competition with AI in Business: How China Differs." The report said 91% of Chinese companies have increased their spending on AI projects since last year. In the U.S., that number is 74%. Within the last three years, 60% of Chinese companies have adjusted their business models to orient their operations around AI, while only 53% of U.S. companies have followed suit.
Much of these changes in China's business realm are driven by the country's government. Last year, the government released details of its national strategy for AI, which calls for investing $150 billion in companies developing AI technologies, with the goal of being the world leader in, among other areas, robotics and AI chip production by 2030. The government intends for this to power an economy with AI deeply embedded in most of its industries, from manufacturing to finance.
The Chinese government's efforts are buttressed by the regime's centralized approach to economic management. Once the government sets a priority, investment follows. In more democratic, capitalist countries, money follows business opportunities.
And while private investors in the West see potential in AI and are putting money into the technology, the tech community is still waiting for transformative business models that would justify the intensive level of investment made by the Chinese government.
The state of American AI investment
That's why the story is so different in the U.S. In 2016, the Obama administration's National Science and Technology Council produced a strategic plan for AI research and development. The national strategy for AI called for long-term investment in AI technology research and development -- both through federal projects and increased support for academic research -- and efforts to train and build an AI workforce.
Pranay Agrawalfounder and CEO of Fractal Analytics
But the plan didn't set out any hard funding targets, and there are few signs that the current administration is acting on the plan's recommendations.
AI research and development has largely fallen to the private sector, which is starting from a robust position. The U.S. is home to world-leading institutions in AI both on the enterprise side -- with companies such as Facebook, Google and IBM -- and in academia.
But these institutions are incapable of keeping up with the pace of global investment on their own. Total spending in the Chinese economy on AI research and development is expected to surpass spending in the U.S. in 2019, according to a report by the U.S. House of Representatives' Subcommittee on Information Technology.
"I think governments need to invest in this area," said Pranay Agrawal, founder and CEO of AI technology vendor Fractal Analytics, based in Jersey City, N.J. "The Chinese government has announced that it will be a priority. They're putting in capital, partnering with the private sector to fuel their ambition at the same time as the U.S. has announced a reduction in research projects."
The risks of falling behind
There are a number of dimensions to falling behind in AI investment. One of the most serious threats comes from a gap in military applications of the technology, as the Subcommittee on Information Technology's report argued. But the problem is more pervasive.
The Chinese government views AI as a foundational technology for the majority of work in the 21st century and is trying to build an economy around it. As more and more work requires people with AI skills, jobs are likely to flow to whatever country can offer the best-trained workforce with the most robust infrastructure. Countries -- including the U.S. -- that fail to build up this kind of national AI infrastructure and, more broadly, set a national strategy for AI, could find themselves at a substantial economic disadvantage in the years ahead.
"The risk is that companies based in the U.S. get left behind," Agrawal said. "If [AI] is the engine for growth, you want to make sure the engine is running well."