Design News is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

AI Steps up to Solve Semiconductor Challenges

Article-AI Steps up to Solve Semiconductor Challenges

Peter Dazeley/The Image Bank via Getty Images AI in semiconductor design
Companies turn to artificial intelligence to close the labor gap in semiconductor design.

The semiconductor industry is growing more competitive. Engineering talent shortages, shrinking time-to-market windows, and never-ending design complexity exacerbate the traditional design processes. These factors have increased the need for improved design results in semiconductor power, performance, and area (PPA). Those who can optimize these metrics faster and at a lower cost improve their competitive position.

One way to tackle these challenges without having dozens of engineers spend months, even years, designing chips is to use AI-enabled tools. We caught up with Arvind Narayanan, senior director of product line management at Synopsys EDA Group, to get his view on emerging solutions in chip design. Synopsys is a chip design company that deploys artificial intelligence to meet design challenges.

Design News: What are the current challenges in the semiconductor industry?

Arvind Narayanan: Moving to smaller geometry is one challenge for semiconductor design. Plus, there’s a new focus on power with the DoD mandate to reduce power. Talent shortages have become a challenge. By 2030 there will be a 30% worker shortage in the semiconductor industry. This brings about the need for greater productivity from existing personnel. How do you improve productivity? Artificial intelligence is one of the drivers.

AI can boost productivity while also improving the quality of the chips. Engineers can use AI in every stage of the chip design process to help discover the impossible – from system architecture, design, and manufacturing, to accessing AI-driven tools in the cloud. AI developers are on a mission to embed intelligence into the world’s leading industries to transform the way we live, work, learn and play. 

SynopsysSynopsys AI stack

DN: What are some of the ways you can use AI in chip design?

Arvind Narayanan: You can use AI to understand the challenges and learn how to meet those challenges. You can apply AI tools, and they will get you to 80% of the design quickly. After that, developers can take it to the finish. You can immediately benefit from artificial intelligence.

AI is a proven solution that enhances engineering productivity and silicon quality, even making things possible that had previously been impossible. All of this comes from AI. We have AI as part of our implementation. We started by applying AI to the usable space. We have now expanded AI to the entire stack.

Gorodenkoff/iStock/Getty Images Plus via Getty ImagesAI in semiconductor design

DN: Explain some of the AI-based tools.

Arvind Narayanan: Engineering ingenuity has led to advancements like AI-powered chatbots, surgery-performing robotics, and self-driving cars. It has also produced solutions that offload repetitive chip design, verification, and testing tasks, allowing engineers to focus on what they do best: innovate.

We’re going to expand AI in all of the domains. There is a lot of work in manufacturing going from analog. The entire stack has to benefit from AI. In manufacturing, you can use the machines you already have and provide results that ate significantly better. Each domain has different slices that need to be solved. You can slice and dice the AI and evolve as you move forward.

DN: Where do you see AI going forward in the semiconductor industry?

Arvind Narayanan: Climate change, viral epidemics, and food insecurity present massive challenges. This has triggered the need for larger, more complex chips to process sophisticated algorithms, whether it’s in edge devices or in servers nestled inside hyper-scale data centers.

Optimizing such vast designs for PPA and ensuring that the chip will perform as intended is a huge undertaking, one that is no longer feasible for humans to take on without a boost in capacity. To get there you need AI.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.