AI enthusiasts must have read this book called “AI Superpower” by Kai-Fu Lee. The author shares his perspectives on how China threatens the US as the global superpower for Artificial Intelligence (AI). He also provides a projection on how AI will change the future world and how humans should prepare for it.
China and AI
He argues that the AI field has been mature enough and its technological trajectory has shifted from invention to the implementation process. China might possess less AI Scientists than the US, but they have many AI Engineers to become the global leader of AI. Those tinkerers did not have to find rocket science, they just had to know enough about how AI worked to turn its power into profitable systems.
As the innovation theory said, “Invention concerns the original development of some novel would-be process of production or product while innovation entails its actual introduction and tentative economic exploitation.” China’s engineers can utilise the already invented AI and innovate it into something useful, equipped with a massive amount of data they collected from Chinese people. The government, which has the ambition to become the world’s leader of AI by 2030, also supported this notion by giving subsidies for AI startups and generous contracts to speed up the adoption of AI technology across the country.
Some samples of practical AI products in China include autonomous drones, pay-with-your-face systems, and intelligent home appliances. Alibaba, one of China’s internet tycoons, has taken the lead on the smart city called “City Brains”, the massive AI-driven networks that optimize city services by drawing on data from video cameras, social media, public transit, and location-based apps.
AI Waves
According to Mr. Lee, there are 4 waves of AI applications, where each of them has a different adoption percentage between the US and China:
- Internet AI = to predict what users want to see/watch/click/buy, as the users are automatically labeling data as they browse the internet. In 2018, the percentage of domination in the market is 50 per cent China and 50 per cent the US, but he predicted China will lead the Internet AI by 60-40 in 2023.
- Business AI = to manage well-structured corporate data sets, such as historical stock prices, credit-card usage, and mortgage defaults. Chinese companies have never truly embraced enterprise software or standardized data storage, instead, they keep their books according to their own systems. Clearly, the US dominated the Business AI applications compared to China by 90-10 in 2018 but projected China will close the gap to 70-30 five years later.
- Perception AI = to recognise the object, face, voice, smart sensors, audio, pictures using smart sensors, augmented, and virtual reality. With the immersive online-merge-offline (OMO) applications in Chinese retails and transportations, such as pay-with-your-face systems, China led the Perception AI by 60-40 in 2018, and will even increase to 80-20 in 2023.
- Autonomous AI = is a mix of all other AI waves, for example, self-driving vehicles, autonomous drones, and robots. In 2018, the US dominated the market with Google, Tesla, and Uber self-driving cars, but soon China will catch up by 50-50 in 2023.
Living in the Future World of AI
Mr. Lee projected around 40-50 per cent of jobs would be replaced by AI in the developed countries by 2030, with white-collar workers hit the most. I believe developing states like Indonesia would last longer until AI reaches the country. But, would we remain safe? There might be a spill-over impact. Those affected people in the developed countries might move and take over jobs in Indonesia. Getting paid less is better than not having jobs at all, isn’t it? (plus the benefits of traveling in exotic provinces).
Moreover, companies in Indonesia would be glad to hire “expatriates” at a low cost compared to hiring local staff. In this regard, Indonesians should worry about AI-induced employment in Indonesia. Not because AI would be used widely here (yet), but rather because of people in developed countries whose job is replaced by AI would come and take over ours. AI would no longer generate a creative destruction process but will make destructive creation.
Silicon Valley entrepreneurs love to describe their products as “democratising access”, “connecting people”, and of course “making the world a better place”. Whereas what they really do is leveraging Information and Communication Technology (ICT) for their huge gains. There is a fear that AI development would divide populations into AI Elite and Useless Class.
Here is the list of policy recommendation provided in the book for adapting to AI economy in the future:
- Reduce work hours (time). Instead of reducing workers altogether, it is suggested to reduce the work hours and make multiple people sharing the same jobs. Although the take-home pay per person might be reduced, the unemployment in general would be minimised. Even Larry Page, the co-founder of Google, phrased his support on changing the culture of five-day work a week into four-day.
- Retrain workers (skills). Encourage people to take lifelong learning via the online educational platforms, and force them to change their career every few years to gain diverse skills
- Redistribute income (compensation). Provide universal basic income (UBI) for all residents, not only jobless people, where the funding comes from steep taxes charged to AI companies
I am personally touched by the author’s story towards the end of the book. All this time he worked hard in the AI field, but then he suddenly struggled with a health issue, and realised that all really matters are family and friends around him that remain there whatever his condition was. As he said, “Mesmerised by my quest to create machines that thought like people, I had turned into a person that thought like machines.”
AI, and other kinds of technology, is a catalyst tool to maximise our impact and change the world for the better. But it should not dictate our life.