Microsoft CEO Satya Nadella warned enterprises using AI about the “reverse information paradox” which could lead to the company having to pay twice — once in cash, once in proprietary knowledge — in a lengthy social post on Sunday.
“In the AI age, the buyer risks giving away knowledge, just to use what they bought. You essentially pay for intelligence twice, once with money, and again with something even more valuable: the proprietary knowledge you must reveal to make that intelligence useful. The better you want the model to perform, the more of that knowledge you have to feed it! Over time, the information asymmetry becomes increasingly skewed. The seller learns more and more about you as you use what you purchased, while you learn very little about what the seller is learning in return,” Nadella wrote on X. The post received nearly 10 million views at the time of this article.
READ: Microsoft CEO Satya Nadella warns of few AI giants ‘destroying entire industries’ (June 15, 2026)
AI models typically learn from user inputs to improve their performance. However, the knowledge they gain is generally retained by the AI company, which can use it to train and improve future models.
“In consuming intelligence, you are creating intelligence. And what you create should belong to you. This is your particular intelligence, in Hayek’s sense: the knowledge of time, place, and circumstance that no one else can hold. It knows what you value, and how you measure success. While the great innovation that comes from model providers having fair use rights to train models on public data is needed, I find it ironic that the status quo is to then turn around and impose restrictive terms of on distillation, and to reserve the right to learn from customer usage and interaction data,” Nadella added.
Nadella also highlighted a few ways through which companies can preserve their competitive edge despite this. Nadella said companies should build their own private systems to evaluate AI, allowing them to decide what good performance looks like based on their own business needs. He also urged organizations to keep control of their internal knowledge—including data, feedback, decisions, and institutional expertise—as well as the outputs generated from their own AI tasks and queries.
READ: Satya Nadella says Microsoft was ready to spend $25 billion after Sam Altman firing (May 15, 2026)
He further recommended creating secure, in-house environments where AI models can be trained or fine-tuned using a company’s real workflows without exposing sensitive information to external AI providers. Finally, Nadella said businesses should avoid relying on a single AI model by keeping the software layer that manages AI tools independent of any one provider.
Nadella said separating the orchestration layer from any single AI model gives companies the flexibility to use the right model for the right task, helping improve efficiency, lower costs, and maintain quality. “Bring these four together and you create your own continuous learning loop (i.e. hill climbing machine) that will allow your AI investments to compound the value of your firm,” he noted.
Earlier this month it was reported that Microsoft has begun reducing its reliance on third-party AI models from OpenAI and Anthropic, replacing them with its own in-house AI models across the Office suite.