CMB Stock News Of The Day 📰🗞️🗞️📈📉

“Anthropic Reportedly Exploring Microsoft’s AI Chips in Major Shift Beyond Nvidia” 🚨🚨🚨

Anthropic is reportedly in talks to use custom AI chips developed by Microsoft, signaling another major shift in the rapidly evolving AI infrastructure race.

According to reports, Anthropic could rent Microsoft’s Maia AI chips through Azure cloud services as companies increasingly search for alternatives to the expensive and supply-constrained GPUs produced by Nvidia.

During the early stages of the generative AI boom, Nvidia’s hardware dominated the market almost entirely, becoming the default choice for training advanced AI models. But as the industry matures, the focus is shifting from training models to running them efficiently at scale — a process known as inference.

That transition is reshaping the competitive landscape.

Inference workloads require enormous computing power as AI companies deploy chatbots, copilots, and agentic AI systems that process trillions of tokens daily. To reduce costs and improve efficiency, major tech firms are increasingly turning toward specialized chips optimized specifically for inference tasks.

Both Amazon and Google have already seen strong demand for their internally developed AI chips, which are designed to compete with Nvidia in specific workloads.

Microsoft, however, has faced delays in rolling out its Maia chip platform and has been working to establish itself as a serious competitor in the custom AI hardware space.

Landing Anthropic as a customer would represent a major validation for Microsoft’s AI infrastructure ambitions.

Anthropic has quickly emerged as one of the most important players in artificial intelligence, backed heavily by both Amazon and Google while competing directly with OpenAI in the race to build advanced foundation models.

If the deal moves forward, it could also encourage other AI companies to diversify away from Nvidia’s ecosystem and explore alternative inference solutions as demand for AI compute continues to explode worldwide.

The broader trend highlights how the AI race is no longer just about building the best models — it’s increasingly about controlling the infrastructure, chips, and cloud networks powering the next generation of intelligent systems.