Author: Satya Nadella, CEO of Microsoft

Translated by: Yuliya, PANews

Editor's Note: This article is translated from the latest insights of Microsoft CEO Satya Nadella. In the article, Nadella explores the future development forms of enterprises under an AI-driven economy. He points out that as the capabilities of AI models continuously evolve, the core challenge for enterprises is no longer just about what digital tools to use, but rather how to build and accumulate "human capital" and "Token capital," which represent the unique AI capabilities of the enterprise. He emphasizes that the intervention of AI will not devalue humans; instead, it will create a powerful compounding effect due to human subjectivity. At the same time, Nadella warns: we cannot repeat the mistakes of early globalization and allow a few AI models to consume all value; rather, society as a whole should work together to create a prosperous "frontier ecosystem," enabling every organization to control its intellectual property and learning loops, thus achieving genuine and lasting win-win situations in the AI era. Below is the original text translation:

Without an ecosystem to support it, no frontier technology can stand on its own.

Recently, I have been deeply thinking about the future of enterprises in an AI-driven economy.

This transformation is fundamentally different from any previous platform changes. In the past, we used digital systems to enhance human capital. Now, for the first time, we can establish a true cognitive loop between humans and digital systems. This is extremely disruptive, as it fundamentally changes our definition of "work" within enterprises.

The real challenge is no longer about how to use certain digital tools or systems, but rather how organizations will continue to learn, build IP, maintain differentiation, and thrive as AI models can continuously absorb human and organizational expertise and commercialize it.

Every company must embark on building what I call "human capital" and "Token capital." Human capital encompasses employees' knowledge, judgment, interpersonal relationships, creativity, and pattern recognition abilities; whereas Token capital refers to the proprietary AI capabilities built and owned by the enterprise.

It is worth emphasizing that human capital does not devalue with the growth of Token capital. On the contrary, it will only become more valuable! I believe that human subjectivity will become the core driving force behind the growth of Token capital. Humans will set grand goals, establish connections across different fields, expand their networks, and identify the most critical patterns. Without human guidance, computing power will merely spin in place.

This means that the real opportunity is not in selecting the best models, but in building a learning loop above the models that generates a compounding effect between human capital and Token capital. You can outsource a task or even a job, but you can never outsource your "learning" ability. The future of enterprises lies in their ability to continuously accumulate and amplify this learning loop between humans and AI.

This requires us to adopt a completely new architectural approach: allowing every enterprise to build an intelligent system that continuously self-improves over time while still firmly controlling its own IP. A company should be able to replace a "generic" model at any time without losing the "company veteran" expertise that has accumulated in its learning system. In the future era, this will be the key "test" to determine whether you truly possess control and digital sovereignty.

Enterprises need to convert their business processes, domain knowledge, and accumulated judgment into AI systems that improve with every use. Private evaluations should accurately capture whether the models are genuinely making progress on crucial business outcomes for the enterprise (not just looking at external benchmark test results!). Private reinforcement learning environments should empower models to grow stronger within the organization's real data trajectories. Such knowledge bases make the institutional memory within the enterprise searchable and enhance the efficiency of Token usage.

This closed-loop cycle will become the new IP for enterprises. I liken it to a "mountain-climbing machine." Unlike most assets, it has a compounding effect. Every improved workflow will generate better training signals, thereby accelerating the accumulation of the enterprise’s unique tacit knowledge. No matter what new independent model capabilities arise, enterprises that build this closed loop early will possess a competitive advantage that is hard to replicate.

What none of us wants to see is a world where every company in every industry hands value over to a few models that consume everything. If all value is concentrated in a very small number of models, the political and economic system simply cannot tolerate it. Society will never allow a future where AI hollowed out entire industries.

Think back to what happened during the first phase of globalization: due to outsourcing, the entire industrial economy was severely hollowed out. The surface GDP figures looked impressive, but the pain of job loss was real, and its consequences are still felt today. We must not allow such dynamics to recur in the AI era—cannot let a few AI systems seize all economic returns while watching the knowledge of entire industries unknowingly commodified.

In my view, our urgent priority must be to build a frontier ecosystem, not just to develop a frontier model; only then can value flow broadly to every company, every industry, and every country. In this ecosystem, every organization can possess a learning loop that carries its institutional knowledge, enabling its human capital and Token capital to continue generating compounding effects.

This is precisely the principle I have upheld throughout my growth: the external value facilitated by platforms should far exceed the value they retain for themselves, allowing every company to continuously innovate on the platform and create its own value.

When all this becomes a reality, enterprises will not only create value for themselves but also inject vitality into the surrounding economies. Employees will see their professional capabilities amplified, their judgment will be integrated into the system, making it replicable and scalable, and the resulting dividends will also benefit the enterprises and communities they belong to.

This is precisely how enterprises create value for themselves and the broader economic body. This is also the stable balance point we should work together to build.

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。