Artificial Intelligence (AI) is a divisive subject, and one prone to extreme forecasts. Some consider it a technology with the potential to radically transform every aspect of human life for the better. Others consider it “a fundamental existential risk for human civilization.” No one disputes that it is a research frontier that will fundamentally shape our future.

Prominent among the AI optimists is Facebook CEO Mark Zuckerberg, who built a Mandarin-speaking app last year in his spare time, and looks set to roll it out as an integral part of his company’s product in the near future. The doomsday quote above comes from Elon Musk, who’s best known for his SpaceX and Tesla startups, but also dabbles in high-speed tube transit and brain-computer interfaces.

Over the last few days, these two stewards of the future have squared off publicly, Zuckerberg defending his rosy picture of AI’s potential boon to humanity in a Facebook Live stream, and Musk promptly swatting that down with a terse Twitter comeback:

I’ve talked to Mark about this. His understanding of the subject is limited.

— Elon Musk (@elonmusk) July 25, 2017

AI is a dense can of worms; dive in here if you want to begin to sorta grasp the big picture. Simply put, the Musk camp (led intellectually by Oxford philosopher Nick Bostrom) holds that superintelligent computers pose a unique threat to human existence (see: the paperclip problem), and the development of AI technology should therefore be approached with extreme caution. The Zuckerberg camp replies with some form of, “Hey, you don’t know that, and AI can do a bunch of stuff that’s useful for society, like weed out terrorists on Facebook.”

These two positions require very different things from lawmakers. Musk’s cautionary position calls for sweeping regulations to inhibit the rise of a malicious superintelligent AI. Zuckerberg’s implies a preference for non-intervention, or possibly outright support through government investment. On his way out of office, former President Barack Obama had a lot to say to Wired magazine about AI, with an optimistic lean to his comments. In an observation strangely adapted from an old Mao axiom, Obama said:

The way I’ve been thinking about the regulatory structure as AI emerges is that, early in a technology, a thousand flowers should bloom. And the government should add a relatively light touch, investing heavily in research and making sure there’s a conversation between basic research and applied research.

Obama’s White House also raised concerns about China’s competitive rise in the AI field, and while his successor doesn’t seem particularly invested in this topic, China certainly does.

Philosophically, it seems clear that China’s current leadership is squarely in the optimistic technocrat camp when it comes to AI research, development and deployment. In a proposal made public last Thursday, China’s State Council — the government’s chief administrative authority — laid out a roadmap for AI development in China that aims to make the PRC world leader in the field by 2030.

And they will spend to get there. Based on official forecasts, the South China Morning Post reports, AI-related industry will account for 26% of China’s GDP by 2030, and the Chinese government plans to pour money into what it sees as a strategic long-term investment. While this may be cause for concern among AI skeptics and military strategists (China’s development of “highly flexible” intelligent missiles is a matter of great interest to Trump’s security apparatus), it’s a boon for China’s tech sector, which is already mobilizing toward a position at the top tier of AI innovation.

We had surpassed human-level Chinese recognition in 2015; happy to see Microsoft also get there for English less than a year later.

— Andrew Ng (@AndrewYNg) October 20, 2016

Perhaps furthest along is Baidu, China’s search behemoth, which opened its second AI laboratory in Silicon Valley earlier this year. Baidu’s speech recognition AI, DeepSpeech, is capable of better-than-human Mandarin transcription and widely seen as a landmark achievement in speech-learning AI.

Baidu faces stiff competition. Their former AI group leader, ex-Googler Andrew Ng, left in March to return to Silicon Valley and do his own thing. One day later, tech giant Tencent (#476 on Fortune magazine’s latest list of the 500 largest companies in the world) poached a top Baidu scientist to head up their own newly formed AI lab. Alibaba, #462 on Fortune‘s list and Tencent’s chief competitor, has developed an AI platform that it’s applying to fields including medicine, manufacturing, e-commerce, and city planning. Didi Chuxing, the ride-hailing app that beat Uber in China and cruised to a $50 billion valuation in April, opened their own Silicon Valley AI lab in March.

And this is just the tip of the spear. The AI boom carries down to the myriad companies in tech hubs in Shenzhen, Beijing and Hangzhou applying AI to products for mass consumption, from smartphone chips to robot dogs to fried chicken.

One must keep in mind, when reading headlines like “China wants to build a $150 billion AI industry,” that China has the underdog’s tendency to broadcast its goals and achievements more loudly than their counterparts in the West. The US is still the global leader in AI research and development, as Andrew Ng and Tencent AI lead Zhang Tong pointed out at a conference last month in Hong Kong.

Also keep in mind that while “applied AI” (the kind of context-specific intelligence that can, for example, tell a Hotdog from a Not Hotdog) has been here for a while, Artificial General Intelligence (AGI), the actually-thinking AI described in the opening voiceover of any good machine-apocalypse flick, is still something that lives in the future. A bunch of AI researchers recently surveyed by Oxford and Yale said we probably have at least a few decades before AGI happens, but their timeline estimates varied wildly.

Whether your own vision of future AI is Spielberg or Matrix in tone, the vast quantities of financial fuel China just announced it will dump on this fire between now and 2030 is reason for us all to pay attention. Keep an eye on Radii for more coverage of trends in AI innovation in China, and watch the space generally to get a sense of how our interconnected world lurches unevenly into the future.

Cover image: Alibaba’s “Hangzhou City Brain” (aliyun.com)