Anchor Capital flags high-bandwidth memory as AI supply chain bottleneck amid hyperscaler capex surge

  • Anchor Capital flagged AI infrastructure as the next phase of the boom, with compute supply lagging demand across chips, memory, data centers, networking, cooling, power.
  • Hyperscalers spent about USD 250 billion in capex in 2024; forecasts point to above USD 700 billion in 2026, over USD 1 trillion in 2027.
  • Bottleneck shifted from 2023 GPU shortages to high-bandwidth memory shortages; AI data centers seen consuming about 70% of premium DRAM output in 2026.
  • Advanced packaging capacity booked through 2026; networks moving from electrical to optical as clusters upgrade from 400 Gbit/s to 800 Gbit/s.
  • McKinsey estimate cited: 124 GW of incremental AI data-center capacity in 2025-2030, requiring USD 5.2 trillion investment; grids seen as a binding constraint.


Disclaimer: This news brief was created by Public Technologies (PUBT) using generative artificial intelligence. While PUBT strives to provide accurate and timely information, this AI-generated content is for informational purposes only and should not be interpreted as financial, investment, or legal advice. Anchor Capital (Pty) Ltd published the original content used to generate this news brief on July 20, 2026, and is solely responsible for the information contained therein.