If an AI bot “is writing essays or Shakespeare or poetry, you can have a 10% to 15% error and your consumer would know the difference and still be wowed by it.” ~ Shankar Krishnamoorthy
Microsoft’s ‘Copilot,’ initially introduced as an aid for coding and email summarization, has now been adapted by Synopsys for chip design. Synopsys is a leading software provider for chip design and now it has collaborated with Microsoft to use the Azure OpenAI system, creating a Copilot tailored for its design tools.
The task of designing chips, which involves arranging billions of transistors on a small silicon piece, is notoriously challenging, expensive, and time-consuming, often requiring years and a large team of engineers.
Early chip design stages involve engineers describing the chip’s functions in a code-like language. Synopsys trained this new system using vast datasets accumulated over its decades in the industry, focusing on achieving high accuracy in the design process.
Bug fixes are one of the most expensive and time-consuming aspects of chip production. Microsoft has begun implementing the Synopsys Copilot in its own chip design teams, which recently introduced Microsoft’s first internally developed data center chips.
Shankar Krishnamoorthy, Synopsys’ design automation group general manager, emphasized the importance of precision in this field, noting that even a minor error rate could result in costly chip defects.
“This technology not only helps us address issues early on, but also reduces the cost of fixing issues at a later stage,” Erik Berg, principal engineer in the chip design verification and validation team at Microsoft.