<div id="readability-page-1">New switching breakthrough: University of Tokyo researchers demonstrated a 40-picosecond magnetic switch with low energy use, potentially reducing AI hardware's heat and power issues. Capacity race intensifies: Applied Materials says chipmakers are repurposing factories and upgrading tools to meet AI-driven demand for advanced logic, DRAM, and packaging. Roadmaps and risks: Intel is developing 10A and 7A nodes, while analysts warn AI chip stocks could face pullbacks despite long-term growth potential.</div>

AMAT

04335

INTC

SOXL

INTW

SOXX

<p>University of Tokyo researchers have developed a 40-picosecond magnetic switch that could lower energy consumption in AI hardware. Applied Materials reports that chipmakers are upgrading facilities to meet rising AI demand for advanced chips. Meanwhile, Intel is working on 10A and 7A nodes, but analysts caution that AI chip stocks may experience pullbacks despite their long-term growth potential.</p>

- Researchers at the University of Tokyo have developed a 40-picosecond magnetic switch that promises low energy consumption, addressing heat and power challenges in AI hardware.  
- Applied Materials reports that chipmakers are upgrading facilities to satisfy the rising demand for advanced logic, DRAM, and packaging driven by AI.  
- Intel is advancing its 10A and 7A nodes, although analysts caution that AI chip stocks may experience pullbacks despite their long-term growth potential.

Semiconductor breakthroughs meet surging AI-driven demand and supply strain