Recently, I took another quick look through the original version of The Infinite Game. Actually, anyone who has studied microeconomics knows that the Chinese translation is not entirely accurate; "Infinite Game" (无限博弈) would be more appropriate, though the current Chinese title is more commercially appealing.
I’m not necessarily recommending this book, even though its readership remains high. In corporate market competition, there isn't such a clear line between a "Finite Game" and an "Infinite Game." First, you survive; then, you survive longer. No matter how lofty a company's long-term vision is, there will always be quarterly and annual KPIs (or OKRs—no matter how nicely you put it, it's just old wine in new bottles). Similarly, a company might just be "scraping by" every quarter, yet it may actually possess more steadfast long-term goals.
In a game, as long as you don't leave the table, you must do everything possible to win; surviving until the end is the ultimate victory. In that sense, Squid Game is perhaps closer to reality.
Of course, as a bestseller, many of the perspectives and viewpoints in the book are still worth referencing. Looking back five years after its publication, the "counter-examples"—specifically the discussions regarding Microsoft—are perhaps more interesting in the present context.
This book was released in 2019. At that time, the mobile internet was reaching its historical peak. The sole reason for criticizing Microsoft then was that it had missed the era of the mobile internet. However, Microsoft caught the cloud wave, with Azure firmly holding the second-largest market share. It may have also grabbed the key to the future: OpenAI. In contrast, Apple has been struggling since the arrival of the AI era.
So, if The Infinite Game were to have a sequel, would the protagonists swap roles? Has Microsoft successfully staged a comeback through AI?
On the surface at least, the answer is yes. This is because Microsoft made a "forward-looking" investment in OpenAI. However, firstly, I believe even if the CEO were still Ballmer (the target of criticism in the book), they likely would have made the same investment decision—especially since Gates and Ballmer still wield massive influence over major decisions despite not being CEO. Secondly, the most important reason Microsoft invested in OpenAI wasn't just that AI is important (tech giants have known this for over a decade), but that their own internal teams were far inferior to OpenAI.
Thus, at least in the high-tech industry, commercial outcomes are often determined by technical factors. Any company, once it grows large, will develop "ailments." Having lived through that history, one knows that Microsoft actually took mobile very seriously—from the early Windows CE and close cooperation with Motorola to the later acquisition of Nokia. You cannot say this wasn't an "Infinite Game" mindset. The problem was simply that Windows, no matter how it was modified at that time, was not suitable for mobile phones.
However, Microsoft's dominance in desktop systems and its solid foundation in enterprise services successfully helped it stage a comeback in the cloud with Azure.
Initially, the immense dominance shown by Google's team in AI also forced Microsoft to place a heavy bet on the other of the then-"twin stars," OpenAI (the other being DeepMind, which Google had already acquired).
It is the trends in technological development, the landscape of technical competition, and the compatibility between new technology and one's core products—rather than commercial competitive decisions—that determine the state of competition at any given stage.
Returning to the question: Has Microsoft staged a comeback through AI? Or rather, can Microsoft return to its peak of dominance through AI?
I remain highly skeptical of the answer, for the same three technical reasons mentioned above.
Since last year, I have been advocating a specific viewpoint triggered by the well-known global events of the past few years. The direction of technological development five years ago versus today follows two completely different paths:
- While "Working from Home" remains controversial, it has greatly stimulated and enhanced the innovation speed and productivity of the most creative individuals, whose behaviors largely dictate future technical trends.
- During the period when everyone was forced online, traditional methods of team collaboration faced massive challenges. Core teams became increasingly miniaturized, and tech stacks were restructured. Fewer programmers are now controlling more technical resources. A series of cloud-native technologies and automated processes have reconstructed previously lengthy workflows.
- ChatGPT was born of the era, but what truly allowed Generative AI to produce such a massive impact in such a short time were the two prerequisites mentioned above: core teams controlling vast technical resources (most importantly, large computing clusters) and a wealth of cloud-native technologies and automated processes that allowed Generative AI to deploy a massive array of tools the moment it launched.
- Because building application demos and prototypes has become extremely easy, the "Demo" has replaced the PPT as the most important foundation for fundraising in this era. Among the roles of programmer, product manager, and salesperson, the individual who can handle the entire process is now the most important.
- As a result, we see more and more "programmers" stepping into the spotlight, becoming "big salespeople" alongside founders or CEOs.
- Behind these symptoms lies a reality: either the people "actually doing the work" define the product, or the "boss" is actively involved throughout the entire process.
"Shortening processes and reducing pure management roles" will still take some time.
Today, we indeed see that the layoff process in large enterprises has not ended. While a global economic slowdown is a factor, the shortening of decision chains brought by technological change has exposed the "negative contribution" of many roles. Big tech companies are even enjoying a "dividend" of technological change where the more they lay off, the better their revenue performance looks.
In fact, during any era of transition, this process of change at the commercial level is similar, though each era has different technical drivers.
A trend that differs greatly from the previous generation of technology is that the previous generation targeted low-cost standardization. Thus, it constantly expanded in scale, standardizing processes and management. From a macro perspective, the increase in decision chains actually reduced long-term costs. That is how more and more "cogs" in management positions were created.
This generation of technology (or the next generation currently beginning) targets low-cost, true personalization (not "one size fits many," but "one size fits one"). It emphasizes faster implementation, shorter decision processes, and a drastic reduction of redundant middle and back offices.
We are now seeing the positive impact on Meta after Zuckerberg optimized the management structure and directly intervened in product R&D. We are seeing Google regain strong competitiveness in AI after Sergey Brin returned to the technical front lines...
As for Microsoft, without the role of OpenAI, the outlook might still be quite negative. And what about the positive example of the "Infinite Game," Apple? From a technical foundation perspective—whether it's chips, operating systems, or the all-in-one ecosystem and imagination—I believe it is currently in the most advantageous competitive position.
However, the lack of a "tech fanatic" or an ultimate "product manager" remains a huge ❓ in my mind.
Returning to reality, in the stage where "increasing R&D investment" is needed to make "AI better," the dominance of tech giants will likely continue to strengthen for at least the next two to five years, rooted in their massive technical accumulation and talent advantages.
Yet, as more and more "independent programmers" form a collective force in pursuit of "true personalization," and as core members leaving big companies can easily secure funding of $100 million or more...
The change has already begun rapidly beneath the surface.