IT Morning Brief 0926: Microsoft Launches New Copilot 'Super App'; Musk Plans 1.1 Million Nvidia GB300 GPUs by Year-End
Summary
Microsoft officially launched its new three-in-one Copilot 'super app' combining chat, coding and agents, saying it aims to define how work gets done in the AI era. Musk disclosed his AI compute plans: Colossus 1 already runs 230,000 Nvidia GPUs, Colossus 2 has 550,000, with 220,000 GB300s coming online next week and another 440,000 by year-end, for a planned total of 1.1 million. Separately, Meta shares rose on the popularity of its Muse product, lifting Zuckerberg to the world's fourth-richest person on the Forbes list.
Bubble analysis
Musk's plan to deploy 1.1 million Nvidia GB300 chips by year-end is direct evidence that the AI compute arms race is still accelerating: massive hardware purchases mean capex keeps ballooning, and whether such spending can be covered by AI revenue is the heart of the bubble debate. But the item gives only chip counts, not dollar figures or matching revenue, so it is strongly related rather than decisive.
Related signals
linked by shared tags
- Nvidia-Backed Firmus Plans Australian IPO at $30 Billion Target Valuation#datacenter#nvidia
- Report: Tencent Leases Oracle Southeast Asia Data Centers — About 100,000 Advanced AI Chips, Deal Valued at Roughly $7 Billion#datacenter#nvidia
- Toshiba Plans to Double HDD Capacity for AI Data Centers, Investing About 60 Billion Yen#datacenter
- Behind the 4 Trillion Compute Network: What Does the Tech Foundation of 'New Quality Productive Forces' Look Like?#datacenter
- Qineng Technology's Datong Super Energy Complex Tianzhen Computing Park Phase I Tops Out#datacenter