Claude goes into production: UST brings Anthropic's AI to chip validation

On July 12, 2026, Anthropic announced a partnership with UST, a company that builds engineering systems for chips, cars and IoT devices. Claude Code will be integrated into UST platforms to automate hardware testing and validation, and 20,000 UST engineers, architects and consultants will be trained to work with the model.

Physical AI: testing built into the process

Physical AI is AI embedded in the engineering processes that produce real-world things. Before a fab is started for millions of chips, engineers stress-test the design. Any error caught early is cheaper: finding a bug during verification is a day's work; finding it after committing to production costs hundreds of thousands in losses. That is when the difference between early and late shows up in its starkest form.

How Claude speeds up chip validation

The key example is UST's iDEC platform, which engineers use to validate hardware and silicon before it goes into production. This is usually manual work: engineers write test scenarios, run them, read the results and repeat the cycle many times. Claude Code now reads chip schematics and pinouts directly, generates and runs regression tests (checks that a design change has not broken anything critical), compares live data from real hardware with its digital twin, and finds firmware problems.

• The iDEC platform with Claude cuts the validation cycle from 4 days to 48 hours
• Speedup of 50–70%: the standard 4-day cycle now takes 2 days
• Claude reads schematics and pinouts, and writes and runs tests automatically
• Engineers do not need to write hundreds of lines of test code by hand
• The system compares live hardware with a digital twin to detect problems early

Expansion into healthcare, telecom and banking

UST also uses Claude in three of its other platforms. CarePath, for healthcare, helps insurers and providers extract insights from fragmented medical data. The system recommends next steps for care teams, and a human approves every action before it reaches a patient.

IntelliOps, for telecom, helps network operators who see thousands of alerts a day, most of them noise. Claude helps find real problems, predict failures in the radio network (towers, antennas) and reduce downtime.

FinX, for banking, works with legacy systems where the ledger is updated once a day. Claude plugs into workflows to automate case processing, servicing and decision support.

Global training and certification

UST has committed to training 20,000 of its employees: engineers, architects, consultants and industry specialists. Anthropic supports this through the Claude Partner Network, providing certification and technical guidance. UST becomes a "Global Premier Partner" of the network.

"Our partnership with Anthropic reflects UST's commitment to helping clients adopt AI with confidence and achieve business results.

By combining Claude with our engineering expertise, we bring solutions that improve performance and speed up outcomes," said Krishna Sudheendra, CEO of UST.

What this means

It signals that AI is moving out of the lab and into production systems serving millions of users and products. Physical AI, meaning neural networks embedded in engineering shops, car assembly lines and banking systems, is becoming reality rather than speculation, and Anthropic is positioning Claude as a tool for this layer of engineering.