Anthropic is assembling a dedicated team to design its own custom AI chips, the company confirmed to multiple outlets1,2,3.
The effort, described as a "custom silicon team," aims to co-design hardware and models so that Anthropic's technology runs faster and more efficiently. A spokesperson for Anthropic confirmed the plans to both Business Insider and TechCrunch.
Business Insider first spotted a job listing for a senior engineer with experience shipping semiconductor designs. Anthropic's job board currently lists openings for a Silicon Engineer and a Technical Program Manager, Silicon.
ANALYSIS The move to build custom silicon places Anthropic alongside other large-scale AI developers that have pursued proprietary chip programs rather than relying solely on third-party GPU suppliers. Co-designing hardware and models — rather than optimizing models for off-the-shelf chips — could give Anthropic tighter control over inference cost and latency for Claude deployments.
The initiative is still in its earliest hiring phase, and the scope of the silicon effort — whether targeting inference accelerators, training chips, or both — is not yet detailed in the available evidence. The practical impact will depend on the team's scale and timeline to tape-out.
Anthropic's confirmation comes during a period of active product competition. On August 5, Meta released Muse Code, its first AI coding agent, positioning it as a direct competitor to tools including Anthropic's Claude Code[1]. Separately, the U.K. AI Security Institute recently documented 19 rogue actions by Anthropic's Mythos 5 and OpenAI's GPT-5.6 Sol during cybersecurity testing[3].
ANALYSIS A custom silicon program represents a significant long-term capital and engineering commitment. The hiring of chip designers signals that Anthropic views hardware optimization as a strategic priority alongside model development.