Microsoft CEO: In the AI era, how do we define a company's moat?
Original Title: A frontier without an ecosystem is not stable
Original Author: Satya Nadella, Microsoft CEO
Translation: Peggy
Editor's Note: Microsoft CEO Satya Nadella believes that in the AI era, a company's true competitive advantage lies not in betting on the strongest model, but in whether it can turn its own workflows, domain knowledge, organizational judgment, and employee experiences into a continuously evolving learning system. In other words, companies cannot just purchase AI capabilities; they must have their own "learning loops" (systems that continuously reinforce human experience, business processes, and model capabilities).
In this framework, future companies will accumulate two types of capital: human capital, which includes employees' knowledge, judgment, networks, creativity, and pattern recognition abilities, and Token Capital (AI capabilities built and owned by the company). Nadella emphasizes that AI will not devalue human capital but will make human goal-setting, cross-disciplinary connections, and critical pattern recognition abilities more important. Without human direction, computational power just spins in place; without the organization's own knowledge base, no matter how strong the model is, it remains an external tool.
The core argument of this article is: a frontier without ecosystem support will not be a stable future. The value of AI should not be swallowed up by a few general-purpose models; instead, it should form a frontier ecosystem, enabling every company, industry, and country to have its own learning loop. Companies need to establish private evaluation, private reinforcement learning environments, and searchable knowledge repositories to transform implicit experiences into reusable, scalable, and iterative system capabilities. The real moat may not be a specific model itself but the fact that even after switching to a general-purpose model, companies will not lose the accumulated "company veteran" experience.
This is also the key to corporate sovereignty in the AI era: whoever can turn organizational knowledge into a system of continuous compounding gains will be able to retain IP in a future of rapid model iteration, amplify employee capabilities, and keep the economic value brought by AI within their own business, industry, and community.
The following is the original text:
I've been thinking recently about what the future holds for companies in an AI-driven economy.
This transformation is unlike any platform shift we've seen before. In the past, we used digital systems to augment human capital; but this time around, we're for the first time able to establish a true cognitive loop between humans and digital systems. This is a very mind-bending thing because it will change how we think about "work" within companies.
The real key question is not how a particular digital tool or system is used, but how organizations continue to learn, accumulate intellectual property, differentiate themselves, and thrive in a world where an AI model can continuously absorb human and organizational expertise and commodify it.
Every company must build what I call Human Capital and Token Capital. Human Capital includes employees' knowledge, judgment, network, creativity, and pattern recognition skills; while Token Capital is the AI capability the enterprise itself builds and owns.
Importantly, as Token Capital grows, Human Capital does not become less important. On the contrary, it becomes even more critical. I believe human agency will be the core driver of Token Capital growth. Humans will set ambitious goals, connect dots across domains, build relationships, and identify truly significant patterns. Without human direction, computational power will merely spin its wheels.
This means that the real opportunity lies not in choosing the best model, but in building a learning loop on top of the model, enabling Human Capital and Token Capital to compound each other. You can outsource a task, you can even outsource a job, but you can never outsource your own learning. The future of enterprises lies in whether they can sustain compounding learning between humans and AI.
This requires a new architectural mindset: every enterprise should be able to build an intelligent organism system that improves over time while retaining control of its intellectual property. A company should be able to swap out a "jack-of-all-trades" model without losing the "company veteran" type of specialized expertise accrued in its learning system. This will be a key test of enterprise resilience and sovereignty in the future era.
Enterprises need to translate their workflows, domain knowledge, and long-accumulated judgment into AI systems that continuously improve with each use. Private evaluations should measure whether the model is truly getting better at the business outcomes the enterprise cares about, not just against external benchmarks. Private reinforcement learning environments should make the model stronger based on real internal trajectories. An enterprise knowledge base will make institutional memory queryable and enhance Token utilization efficiency.
This loop will become the new intellectual property of the enterprise. I see it as a "climbing machine." And unlike most assets, it will compound. Every workflow improvement will yield better training signals, accelerating the accumulation of the enterprise's unique tacit knowledge. Companies that establish this system earlier will gain an advantage that is hard to replicate, regardless of how individual model capabilities evolve in the future.
What we least want to see is a world where every company in every industry cedes value to a few models that devour all available content. If all value ends up captured by a few models, the political-economic structure would never tolerate such an outcome. An AI future that hollows out entire industries cannot obtain societal permission.
Think about what happened in the first phase of globalization: the entire industrial economy was hollowed out through outsourcing. While GDP numbers may have looked decent on the surface, the real industrial shift and employment impact did occur and its consequences are still being felt today. We cannot bring this dynamic into the AI era—allowing a few AI systems to capture all economic gains while the collective knowledge of entire industries is commoditized and hollowed out beneath them.
In my view, our priority must be to build a cutting-edge ecosystem, not just a cutting-edge model. Only then can value flow widely to every company, every industry, and every country. In such an ecosystem, every organization can have its own learning loop, encode its institutional knowledge, and allow human capital to exponentially compound alongside tokenized capital.
This is also the platform ethos I have always believed in: the value created on top of a platform should be greater than the value captured by the platform itself; every company should be able to innovate continuously and create its own value.
When this is achieved, companies will create value for themselves and for the economic environment in which they operate. Employees' expertise will be amplified, their judgment will become part of the system, becoming replicable and scalable, with these benefits flowing back to the company and its surrounding community.
This is the way for companies to create value for themselves and for the broader economy. It is the stable equilibrium we should collectively strive to build.
Welcome to join the official BlockBeats community:
Telegram Subscription Group: https://t.me/theblockbeats
Telegram Discussion Group: https://t.me/BlockBeats_App
Official Twitter Account: https://twitter.com/BlockBeatsAsia
