Anthropic Explores Building Custom AI Chips Amid Surging Demand for Claude Models

The AI company is in early discussions to develop its own silicon as it grapples with a global chip shortage.

Apr. 10, 2026 at 6:20am

Got story updates? Submit your updates here. ›

A highly detailed, glowing 3D illustration of a complex circuit board with intricate patterns of neon cyan and magenta lights, representing the advanced hardware infrastructure powering Anthropic's AI models.Anthropic's exploration of custom AI chip development aims to bolster its hardware capabilities and meet the surging demand for its advanced language models.Washington Today

Anthropic, the parent company of the popular AI chatbot Claude, is reportedly in the preliminary stages of evaluating in-house chip development as it faces a global shortage of AI chips and surging demand for its models. The effort remains exploratory, with no formal commitment or dedicated team in place yet, but the company could still opt to continue purchasing chips rather than building its own.

Why it matters

The development reflects a broader trend across the AI sector, as designing advanced AI chips is both complex and costly, with estimates suggesting development can run as high as $500 million. Anthropic's deliberations come amid accelerating demand for its Claude models, with the company saying its annualized revenue run rate has surged past $30 billion, up from roughly $9 billion at the end of 2025.

The details

Anthropic is reportedly in the preliminary stages of evaluating in-house chip development, with no formal commitment, finalized design, or dedicated team in place yet. The company could still opt to continue purchasing chips rather than building its own. The discussions come amid a global shortage of AI chips required to power and advance next-generation AI systems, as well as a legal controversy with the Pentagon over the company's refusal to ease safeguards on its Claude chatbot for uses such as surveillance or autonomous weapons.

  • Anthropic earlier this week said its annualized revenue run rate has surged past $30 billion, up from roughly $9 billion at the end of 2025.
  • Earlier this week, a federal appeals court in Washington, D.C., declined to temporarily block the Pentagon's decision to label Anthropic a national security risk.

The players

Anthropic

The parent company of the popular AI chatbot Claude, which is exploring in-house chip development amid a global shortage and surging demand for its models.

Pentagon

The U.S. Department of Defense, which has labeled Anthropic a national security risk after the company refused to ease safeguards on its Claude chatbot for uses such as surveillance or autonomous weapons.

Got photos? Submit your photos here. ›

What’s next

Anthropic's chip development efforts remain in the early stages, and the company could still decide to continue purchasing chips rather than building its own. The legal controversy with the Pentagon is also ongoing, with the federal appeals court declining to temporarily block the Pentagon's decision to label Anthropic a national security risk.

The takeaway

Anthropic's exploration of in-house chip development reflects the growing challenges faced by AI companies as they grapple with a global chip shortage and surging demand for their advanced models. This trend is not unique to Anthropic, as the AI sector as a whole is shifting towards custom silicon development to meet their specialized computing needs.