Chip verification, one of the most time-consuming steps in semiconductor development, is getting an artificial intelligence overhaul. Siemens has partnered with Nvidia to accelerate the process of confirming that complex AI chip designs actually work as intended before they go into manufacturing, according to Evertiq.
Verification is not glamorous work, but it is essential. Before any chip reaches production, engineers must simulate and check millions of possible operating conditions to ensure the design behaves correctly. As chips grow more complex to meet the demands of AI workloads, that verification process has grown longer and more expensive. It is now considered one of the primary bottlenecks in bringing new chips to market.
Siemens, through its electronic design automation business, is applying Nvidia's accelerated computing technology to compress that timeline. The approach uses GPU-based simulation to run verification tasks far faster than traditional CPU-based methods allow. The goal is to let chip designers iterate more quickly and catch errors earlier in the development cycle, when fixes are cheaper and less disruptive.
The partnership reflects a broader pattern in the semiconductor industry. As demand for AI-capable chips has exploded, the pressure to shorten design and verification cycles has intensified. Chipmakers and their tooling partners are racing to modernize workflows that were built for a slower era.
That pressure is visible across the industry. Marvell Technology, for instance, recently secured a significant contract to design custom AI chips for Google, a deal that has drawn attention from investors and analysts trying to gauge how the custom silicon market will develop. As reported by Simply Wall St., the Google win raised both growth expectations and valuation questions for Marvell, given the capital intensity of custom chip development and the uncertainty around how quickly those programs scale into revenue.
Together, the Siemens-Nvidia collaboration and Marvell's push into custom AI silicon illustrate where the semiconductor industry is placing its bets. Speed matters. The ability to design, verify, and manufacture AI chips faster than competitors has become a strategic priority, not just a technical one. Governments, cloud providers, and defense contractors are all competing for access to advanced chips, and the companies that can compress development timelines stand to benefit most.
For Siemens, the Nvidia partnership extends its position as a provider of design infrastructure rather than chips themselves. It is a quieter role than building the chips, but verification tools touch every design that goes through the pipeline, giving Siemens broad exposure to the AI chip buildout without taking on the manufacturing risk.
