Written by: Techub News Compilation

Introduction

Artificial intelligence (AI) has evolved from a science fiction concept into a core force reshaping the global landscape. In this profound transformation, former Google global vice president and founder of Innovation Works, Kai-Fu Lee, offers a highly visionary analysis in his book "AI Superpowers," drawing on his unique perspective across top tech companies in China and the U.S. This article interprets his core ideas in depth and summarizes Kai-Fu Lee's systematic thinking on the AI development wave, the nature of Sino-U.S. competition, potential social risks, and humanity's response strategies. This discussion is not only about technological trends but also a profound inquiry into the direction of civilization.

Summary

• The core driving force of AI development has shifted from "discovery" to "implementation," from "expert knowledge" to "massive data," granting countries with advantages in data and application scenarios (such as China) new competitive opportunities.
• The paths of China and the U.S. in AI development are vastly different: the U.S. leads in foundational research and breakthroughs, while China, through gladiatorial-like fierce competition, a mobile-first super application ecosystem, and strong government promotion, has developed unique advantages in application implementation and data accumulation.
• While AI brings enormous productivity improvements, it also leads to severe employment impacts, exacerbating social inequality and posing monopolistic risks of "winner-takes-all," potentially giving rise to a "useless class" marginalized by technology.
• The future of technology is not predetermined; human choices are crucial. We need to innovate beyond simple technological solutions, addressing values, economic models, and social contracts to build an AI era centered on human care and connection.

A Turning Point for AI: From Games to Reality

Kai-Fu Lee points out that a landmark psychological turning point in the field of artificial intelligence was in 2016 when AlphaGo defeated world Go champion Lee Sedol. This victory transcended a mere game; it sent a clear signal to the world: AI's capabilities had undergone a qualitative leap. The engine behind this is deep learning.

Deep learning is fundamentally different from traditional programming logic. It no longer relies on programmers to create countless "if-then" rules for each specific task but instead discovers patterns and rules by inputting massive amounts of data (such as millions of images of cats and dogs) for machines to learn from autonomously. This "data-driven" learning approach has enabled AI to reach or even surpass human levels in specific tasks like image recognition, speech understanding, and credit approval. This means that AI has transformed from the sci-fi promise of "always one step behind" to a reality that has arrived and is infiltrating various industries, bringing unprecedented efficiency improvement potential and planting the seeds for the disruption of existing employment structures.

Two Major Paradigm Shifts: Redefining the Rules of Competition in the New Era

Kai-Fu Lee believes that AI development is currently undergoing two fundamental paradigm shifts that redefine global competition rules.

The first is the shift from the "Age of Discovery" to the "Age of Implementation". The heroes of the Age of Discovery are a small group of foundational researchers primarily concentrated in North America, who laid the theoretical foundations for deep learning. However, after achieving these significant breakthroughs, equivalent theoretical leaps have not frequently emerged. Kai-Fu Lee argues that the focus has now shifted to how to combine existing powerful AI models with vast real-world data to solve specific application problems. The core intellectual breakthroughs have temporarily concluded, and the phase of engineering and data integration has begun.

The second is the shift from the "Age of Experts" to the "Data Age". In the implementation stage, data has replaced algorithm elites as the crucial winning factor. Kai-Fu Lee emphasizes that once basic computational capabilities and engineering talents are available, the scale and quality of data often prove to be more decisive than the sophistication of algorithms. A model constructed by slightly less skilled engineers but trained on a large, high-quality dataset is very likely to outperform a top-tier algorithm trained on limited data. This transformation creates a structural advantage for entities (whether companies or nations) that can generate and obtain massive, diverse, high-quality data.

China's "Gladiator" Advantage: Competition, Data, and Super Applications

Kai-Fu Lee analyzes why the aforementioned paradigm shifts may give China a unique or even decisive advantage in AI applications. This is driven by several mutually reinforcing factors.

Firstly, there is a gladiatorial-esque competitive environment. China's internet battlefield is described as a "gladiatorial arena," with competition raging far beyond that in Silicon Valley. Here, survival relies on extreme speed, continuous micro-innovation, and ruthless execution. Imitating successful models (often regarded as "cloning" by the West) is seen as a practical starting strategy, rather than a sin. The real test is whether one can survive and win through localized innovation and efficient execution amid subsequent brutal market clashes. Meituan founder Wang Xing is a typical example: starting by imitating Groupon and Twitter but ultimately becoming a dominator in the field through extreme adaptation to the Chinese market and relentless competition.

This intense competition unfolds within a mobile-first, highly integrated "parallel digital universe". Chinese users handle everything through smartphones (digital wallets), from socializing and making payments to hailing rides and shopping. This generates an incredibly rich data flow intricately linked with physical life—comprising not only clicks and likes but comprehensive portraits of whereabouts, consumption habits, and lifestyle rhythms. Kai-Fu Lee believes that this data is far more valuable than the primarily online behavioral data used in the West for training AI aimed at interacting with the real world.

The mobile payment revolution driven by WeChat and Alipay serves as the foundation of this ecosystem. China skipped the credit card era, directly entering mobile payments, rapidly popularized through culturally creative features like "WeChat Red Packets." The seamless payment experience has led to explosive growth in O2O (online to offline) services such as food delivery, bike sharing, and ride-hailing, further generating massive amounts of real-time offline behavioral data.

In addition, the Chinese government has played an active role as a promoter, regarding AI as a strategic core for economic growth and global influence. Through the establishment of innovation parks, provision of subsidies, and creation of government-guided funds, a "top-down" planning approach accelerates the construction of an innovation ecosystem, contrasting sharply with the "bottom-up" organic growth model of Silicon Valley, but proving to be an effective catalyst in China's developmental phase.

The Four Waves of AI and Sino-U.S. Competition

Kai-Fu Lee depicts the evolution of AI using the "Four Waves" framework: 1. Internet AI (based on online behavioral data, such as recommendation systems); 2. Business AI (optimizing enterprise processes); 3. Perceptual AI (endowing machines with vision and hearing, such as facial recognition); 4. Autonomous AI (machines capable of decision-making and actions, such as autonomous driving).

He believes that China has a significant advantage in Internet AI (such as ByteDance's personalized recommendations), has made rapid advancements in perceptual AI (such as Megvii's facial recognition), and is leveraging its powerful manufacturing foundation to promote related hardware development. In the most challenging autonomous AI field, the U.S. still maintains its lead through top-tier foundational research, but China is quickly catching up through massive investments and infrastructure development.

In terms of globalization strategies, the paths of the U.S. and Chinese giants differ: American companies often attempt to directly replicate their global models into new markets, while Chinese companies prefer to "arm local insurgents", empowering local players through investment and technology rather than operating themselves.

The Dark Side: Employment, Inequality, and the Risk of a "Useless Class"

Kai-Fu Lee takes considerable space to seriously explore the social impacts that AI may bring, particularly the massive displacement of job positions and increasingly aggravated inequality.

He dismisses the blind optimism that "technology will always create new jobs" (known as the "Luddite fallacy"), pointing out that AI, as a general-purpose technology, with its software attributes, can spread and iterate at an incredibly fast pace, potentially far exceeding the pace of economic structural adjustments and new job creation. He mentions the U.S.'s phenomenon of "decoupling"—where productivity growth coexists with stagnant wages—and warns that AI may exacerbate this trend.

The data feedback loop will lead to "winner-takes-all" monopoly: companies with more data can train better AI, attract more users, generating more data, creating insurmountable barriers. This could engender a future dominated by a few tech giants, coexisting with a large "useless class" of people rendered unemployed or underemployed due to skills becoming outdated, much like the scenario depicted in the science fiction novel "Folding Beijing." Unemployment is not just an economic issue but will erode human dignity, sense of purpose, and mental health.

Insights from Cancer and a Human-Centered Future Blueprint

Kai-Fu Lee shares his personal experience of battling lymphoma (initially misdiagnosed as stage four), which became a pivotal point in his thought transformation. This life-and-death trial awakened him from a past nearly machine-like, "optimization" life focused on maximum impact and achievement, realizing he had neglected family, love, and human connections—things that truly matter.

This profound realization directly shaped his thinking on addressing AI challenges. He believes that we cannot merely rely on technology to fix things or passively accept; we must actively construct a human-centered AI era. He proposed multi-dimensional solutions:

• Create “humanity-veiled” jobs: Let AI handle optimization and routine tasks while humans focus on jobs that require empathy, care, and social connection, such as compassionate caregivers working alongside AI diagnostic tools.
• Promote corporate social responsibility and impact investing: Encourage companies to transcend profit maximization and actively create social value.
• Envision a new social contract: Governments need to play a core role. He proposed the idea of a "social contribution allowance," going beyond unconditional basic income (UBI) that rewards socially beneficial but traditionally unpaid labor, such as caring for the young and elderly, community service, and lifelong learning, redefining "value."
• Advocate for global learning and collaboration: No single culture has all the answers. We should learn from the strengths of various countries, such as the social-emotional education of the U.S., the craftsmanship spirit of Switzerland and Japan, the tradition of volunteer service in the Netherlands and Canada, and the culture of filial piety in China, to collectively explore social models that adapt to the AI era.

Conclusion: The Future is in Our Choices

Kai-Fu Lee's discussion ultimately points to a hopeful yet urgent conclusion: the future of AI is not predetermined. It will be shaped by our choices today—as individuals, entrepreneurs, and policymakers. Technology itself is a tool, and its morality depends on the humanity and institutions of those who utilize it. While we strive to develop powerful AI, we must always be vigilant not to lose the most cherished human qualities: love, empathy, and connection. This book not only analyzes the rise of AI superpowers but also serves as a call to action, guiding us to contemplate "what kind of world we want to create using AI."

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。