- Palihapitiya predicts AI competitors will converge across the full stack, intensifying competition between silicon, cloud, and application companies
- Microsoft introduced its Maia 300 chip, claiming 40% more tokens per watt than leading GPU-based systems
- Elon Musk argues that scaling power, construction speed, cooling, networking, and batteries are the true lasting constraints
- Retail sentiment on MSFT and TSLA is bearish, while SPCX sits neutral despite strong year-to-date performance in MSFT
Venture capitalist Chamath Palihapitiya on Monday underscored power and physical infrastructure as potential lasting constraints in the AI race, following Microsoft‘s announcement of its own silicon chips. He wrote on X that silicon chip companies are moving up the stack while application companies move down, meaning AI is increasingly a space where “everyone competes with everyone.”
He questioned what constraint could ground long-term markets and margins: distribution, a data moat, or a physical factor such as power. Microsoft CEO Satya Nadella responded by calling the company’s silicon push its “biggest bet yet on our own silicon.”
The Maia 300 chip yields 40% more tokens per watt than leading GPU-based systems, while the Maia 200 already serves Teams and Copilot in the US, Europe, and Australia. Elon Musk weighed in on Palihapitiya’s question, saying that “scaling power harnessed, along with super-fast construction, cooling, networking and batteries” is the hardest part.
On Stocktwits, retail sentiment around MSFT and TSLA was bearish at press time, while SPCX was neutral. MSFT has gained 12% year-to-date, TSLA has fallen 16%, and SPCX trades near $163, above its $135 IPO price but below its all-time high of roughly $226.
✅ Follow BITNEWSBOT on Telegram, Facebook, LinkedIn, X.com, and Google News for instant updates.
