ChipsThe story, in brief

IBM is developing a dual-architecture chip to run Arm-native AI apps on Z mainframes and Linux servers

IBM's dual-architecture processor could reshape how enterprises run AI workloads on mainframes—the systems that still process 87% of global transactions.

Paper-cut illustration of an amber microchip with circuit paths extending into a row of data-center cabinets.
The infrastructure powering AI.AI illustration by KeyNews
The KeyNews take

Why it matters

IBM is bridging a critical gap: enabling Arm-native AI applications to run natively on Z mainframes and LinuxONE systems. This matters because enterprises have trillions in legacy mainframe compute; a chip that lets them deploy modern AI stacks without replatforming could unlock massive capital efficiency and shift the economics of AI in the enterprise data center.

The key facts

6 to know
  1. Announced at Hot Chips 2026 conference (August 24, 2026)

  2. Partnership between IBM and Arm Holdings

  3. Dual-architecture design (Arm-native and Z/Linux support)

  4. Target platforms: IBM Z mainframes and LinuxONE systems

  5. Chip is still unnamed and in development phase

  6. Designed to run Arm-native AI applications on legacy enterprise systems

Go to the source

SiliconAnglesiliconangle.com

Publisher excerpt: IBM Corp. said today at the annual Hot Chips 2026 conference that it’s developing a groundbreaking new dual-architecture processor in partnership with the British chip design firm Arm Holdings Ltd. that’s destined to power its next-generation IBM Z mainframes and LinuxONE systems. The new chip,…
Read original report
Back to today's editionMore chips news

Keep reading

Related stories

More from Chips