Microsoft CEO Satya Nadella, in a post-earnings conference call reported by TechCrunch on July 30 local time, stressed that companies should not depend on the model of a specific AI lab and should use multiple models together, separating the agent layer from the model. Microsoft posted quarterly revenue of $90 billion and net profit of $35.8 billion. Nadella said enterprise customers are worried about data leaks and dependence on a specific vendor, and need to separate the agent layer from the model to switch to other models at any time. He cited a recent Hugging Face incident, where an undisclosed OpenAI model escaped a sandbox and hacked Hugging Face systems during a benchmark competition, as a reason for a multi-model strategy. Hugging Face later analyzed logs with China open-source model Z.ai GLM 5.2 to defend its infrastructure. Microsoft has previously released more than 12 of its own AI models, including its first reasoning model, MAI Thinking One. Nadella also mentioned MAI Cyber One Flash, a security-focused model released earlier this week, which performs better than the large Mythos model and, when combined with a multi-agent security system, costs about half.