Having invested in the OpenAI initiative ChatGPT, Microsoft Corporation is actively implementing relevant developments in various fields of activity, and Synopsys, with the support of this software giant, has managed to create a productive artificial intelligence system that helps engineers design new processors faster and look better. for the flaws within them.

The company Synopsys, which produces software tools for developing chips, reported this week that, coincidentally, Microsoft’s Azure Cobalt 100 and Maia 100 processors for cloud systems were also created with the participation of an adapted version of the Copilot assistant. Synopsys specialists who trained this system were faced with the task of achieving a minimum level of errors in the work of this assistant, since the presence of defects in the field of processor development is very expensive for companies. They often spend years and hundreds of millions of dollars to develop a single processor, as well as employing thousands of engineers. All design decisions must provide 99.9% accuracy.

Another problem in developing modern processors with billions of tiny transistors is finding possible flaws and debugging the design. The adapted version of Copilot copes with this task, speeding up the search for problem areas. The generative intelligence system in the early stages of design allows engineers to formulate a technical task in precisely defined expressions, similar to a programming language.

Future artificial intelligence offers some typical design solutions, taking into account the experience that Synopsys experts have accumulated over more than ten years. Such an assistant runs on the computing power of Microsoft Azure OpenAI. Perhaps after further testing, such development tools will be available not only to Microsoft but also to other Synopsys customers.