Microsoft says its own chip leapfrogs those made by other cloud providers like Amazon and Google. But it needs more.

Physical AI marks a transition from robots as programmed tools to robots as adaptable collaborators. That transition will unfold over years rather than months, but the foundation models emerging from Microsoft, Nvidia and Google establish the architectural patterns that will shape enterprise robotics for the next decade.

Microsoft just rolled out Copilot upgrades that make it easier to pick up old chats, handle longer prompts, and use Copilot on Mac and iPhone. Some features arrive in waves, starting with memory in the US.

With Microsoft spending many billions on data centers, and rumors that no one is using its AI, CEO Satya Nadella shared some usage numbers.

The bull case is that Microsoft's significant spending to build out its cloud computing business will eventually accelerate the company's ability to convert its swelling commercial RPOs into revenue, and that this revenue stream will be highly profitable.

Microsoft's capital expenditures, viewed as a barometer for artificial-intelligence spending, may have topped a milestone figure in the latest quarter. The company hasn't spent upwards of $20 billion on capital expenditures in any quarter going back at least to the end of 2021,

Microsoft CEO Satya Nadella talks Copilot, Azure, AI, agents, OpenAI and more on the company’s Q2 earnings call.

In its fourth quarter earnings report, Meta said capital spending plans for 2026 should be in the range of $115-$135 billion, well ahead of consensus forecasts.

Microsoft's capital expenditures surged 66% in the December quarter as the company rushed to spend on artificial-intelligence infrastructure. The company reported $37.5 billion in capex, inclusive of assets acquired through finance leases.

With its newest AI chip, Microsoft plans "wider customer availability" after its initial product was only used for internal projects.