TECHNOLOGY

Anthropic Hires Google TPU Founder Amir Salek to Lead Custom AI Chip Development

PUBLISHED Aug 22, 2026, 12:50 AM ET

Read, Watch or Listen

Anthropic Hires Google TPU Founder Amir Salek to Lead Custom AI Chip Development

Media Bias Meter

Sources: 31

Center 100%

Sources: 31

Artificial intelligence startup Anthropic has hired Amir Salek, the former head of Google's custom silicon infrastructure program, to lead its newly formed custom chip division. Salek, who spent nine years at Google directing Tensor Processing Unit development from version one through four, will oversee Anthropic's internal hardware strategy. The executive move aims to build proprietary semiconductors for training and running large language models, reducing long-term reliance on merchant silicon providers. Prior to Google, Salek spent eight years at Nvidia establishing system-on-chip design organizations. Most recently, he served as a senior managing director at Cerberus Capital Management. The appointment represents a major escalation in Anthropic's infrastructure expansion, following capacity agreements with AMD and Google. Industry analysts note that custom chip development typically requires over eighteen months to yield production hardware, leaving existing supply contracts with Nvidia, AMD, and Google Cloud unaffected. Anthropic has not released an official corporate statement confirming executive titles or start dates.

By Michael Grant | JQJO News

Timeline of Events

On 2013, Amir Salek joined Google to found custom silicon infrastructure program
On March 2022, Amir Salek transitioned to senior managing director at Cerberus Capital
On August 5, 2026, Anthropic formally established an internal custom silicon design group
On August 21, 2026, Bloomberg reported Anthropic hired Amir Salek for custom chips
On August 22, 2026, Financial markets reacted to strategic semiconductor talent acquisition moves
On coming months, Anthropic will scale internal hardware engineering teams significantly
On coming year, initial architectural blueprints for proprietary AI accelerators will emerge
On future quarters, supply chain diversification efforts will expand across global foundries
On long-term horizon, custom silicon prototypes will begin internal model training tests
On production phase, custom chips will enter racks to power Claude models

News Intelligence

Immediate US impact: Tech sector stocks and semiconductor supply chains experience strategic shifts
Possible long-term US impact: Frontier AI labs reduce dependency on traditional merchant silicon vendors
Most affected groups: Semiconductor manufacturers, artificial intelligence developers, cloud computing providers, and tech investors
Reader priority: Monitor official corporate announcements and verified industry trade reporting closely

Media Bias

Articles Published:: 31
Right Leaning:: 0
Left Leaning:: 0
Neutral:: 31

Explain Framing

Left: Emphasizes corporate power consolidation and market competition dynamics Center: Focuses strictly on factual reporting of personnel movements and technological milestones Right: Highlights capital allocation efficiency and private sector innovation independence