Semiconductor Growth from AI-Driven Design Productivity

There’s a lot of pessimism right now, from politics to economics, employment, health, and the threat of conflicts. This pessimism is reinforced by a constant stream of negative news reporting. If you listen to news shows long enough, you’ll feel sure that the bottom is about to fall out. But that would be an overreaction.

There are many very good things happening, and there are plenty of reasons to believe that the world will be much better in the near future. One of the biggest reasons is that we’re learning how to harness the capabilities of artificial intelligence (AI) to work for us and elevate our lives. Allowing AI to take over mundane work will let us humans focus on more important endeavors.

There is a fear that AI will take jobs away. But that’s like worrying about the invention of the wheel taking away jobs. Yes, the job of carrying things was transformed, and the work of carrying things evolved. However, the great benefits of the wheel transformed society, extending the distance things could be carried and increasing the number of different things that could be carried. This created many more jobs, not fewer.

Opportunities

While AI will be used to implement jobs with mundane, repetitive tasks that can be automated, it will create entirely new areas for employment. These new areas should pay more and be more intellectually rewarding than the redundant automated jobs. These new jobs will use AI to enhance the jobs we already have. For example, if AI is used to evaluate documents during a financial audit, it might be able to spot issues that have been overlooked as people slowly peruse those documents. If done well, AI can help managers find patterns in the data that might not otherwise be found. But we will still need humans—and maybe more humans than ever before—to deal with these issues.

In manufacturing, AI can open up new jobs to help ensure data quality from process monitoring systems, with additional jobs focused on data analysis and optimization. More jobs will be created for installing the new systems and improving processes.

Think about exciting new areas for AI to help our society. Imagine that it could be applied to education so the best techniques for learning can be applied to help each student learn at their own pace. Yes, we will still need teachers. But if they had help monitoring the computers that their students use in the classroom, they could be much more productive and improve school test scores. The possibilities of AI-assisted learning are tremendous.

In our industry, the design of new electronic products has become very challenging because the technologies and functionality are so complex. Any mistakes made in the design phase are difficult to discover before the products enter fabrication—and very expensive, if not almost impossible, to fix after the products are manufactured. Applying more engineers to the verification and testing is not the solution—those engineers are much more productive designing new products rather than doubling down on testing. And there just aren’t enough engineers.

The solution is to apply AI techniques to speed design, thereby simplifying the design process. Simplified, AI-enhanced techniques are now being employed in the design of advanced semiconductors, the brains inside the electronics, and the packaging of new electronic systems. By reducing the design difficulty, productivity increases, and the need for extensive testing decreases. This means we can keep up with the increasing need for design complexity by designing more intelligently and with more innovation.

AI for Electronics Design

The increase in the use of AI in electronic design has created many employment opportunities for engineers to develop these AI-enhanced tools and train engineers on how to use them. Employment at Cadence and at other major design companies has increased significantly as these new AI techniques have been employed. And there are so many more opportunities for AI assistance in design that this trend should continue.

We have only just begun to employ these techniques and see the benefits of using AI to help simplify and ensure the quality of electronic design processes, from the design of chips to the design of boards and to the design of entire systems. This has already resulted in the ability of existing design teams to keep up with the growth in complexity, as well as a reduction in the verification time required to ensure the products will work as intended. Indeed, verification has been a great area for AI enhancements and productivity improvements. It’s much better to find out that there’s a problem as the product is designed rather than after it is in the marketplace.

With faster, more accurate design techniques, we are enabling a whole new world full of products that can meet local market needs. Before, because products were so expensive and time-consuming to design, one product had to meet the needs of many countries and groups of people. Now, there’s a promise that, as we continue to gain from AI-enhanced design, it will be easier to create derivatives or entirely new products for local needs. Then we will create even more jobs, close to local, unique requirements and opportunities. Local engineers will be able to hyper-customize solutions to meet their specific needs. All because AI techniques will significantly simplify the design, verification, and testing processes. AI techniques are also being applied to the manufacturing process, further speeding this trend toward localization.

All of these factors will result in an expansion of design starts, which will also mean continued full and growing employment in the existing semiconductor and electronic products value chains. And it means that the world will be better, with new products that meet local requirements, hopefully making life easier and more productive.