Microsoft chief executive Satya Nadella has cautioned that companies putting all their trust in one proprietary AI model are taking an existential gamble.
Speaking on CNN’s “Fareed Zakaria GPS,” Nadella argued that organisations must keep firm control over their own AI infrastructure and the data flowing through it.
His warning was blunt. “Any firm that doesn’t have this control, I will claim will not remain a firm because you’ve essentially outsourced your thinking,” he said.
Nadella laid out a set of practical steps for businesses. He suggested retaining metadata from every model interaction so companies can eventually train their own systems, and deploying AI gateways that keep prompts separate from the underlying models.
He also warned against getting locked into any single vendor’s coding tools. “By keeping the harness separate from the model and the context and memory separate from the model, you absolutely can use multiple models,” he explained.
The comments are notable given Microsoft’s own heavy stakes in the space, including investments in both OpenAI and Anthropic. The advice carries a self-serving edge, yet it lands amid a broader shift toward open-source alternatives among enterprises wary of vendor dependence.
Source: TechCrunch.
Comments 0