What’s next in the world of semiconductors?

Four industry experts weigh in at a semiconductor symposium

There are many perspectives on exactly how the next revolution in the semiconductor space will look. Many acknowledge that the future will be largely impacted by technological advances, such as artificial intelligence and neuromorphic architectures. The challenge lies in predicting exactly how these technologies will develop over the next few years and, as a result, shape hardware development.

At the 2019 SEMI’s Industry Strategy Symposium, four high-profile experts offered their thoughts through an open panel discussion. The participants included Federico Faggin, designer of the first microprocessor, Terry Brewer from Brewer Science, Sanjan Natarajan from Applied Materials, and Michael Mayberry from Intel.

Untapped CPU Potential

Although computing power has improved drastically since microprocessors were first developed, so much potential still remains. We have a long way to go in terms of developing cost-effective, energy-efficient CPUs that can handle the types of projects that scientists and developers are taking on today.

Manage your supply chain from home with Sourcengine

According to Federico Faggin, the key to jumpstarting the next wave in semiconductor innovation lies in understanding how living organisms process information.

“So to me, the next step is not going to come from silicon. It’s moving from our way of understanding systems to an organism, which is a living thing where material flows in and out of a cell… I urge you to look beyond the way we do things. The real stuff is living systems.”

Artificial Intelligence a Necessary Disruptor

Despite gradual improvements in semiconductor and transistor technology, the materials space will fall short of processing needs, should the current trajectory continue. Materials suppliers have been able to rise to market demands over time, however, a disruptor is needed to really propel the space forward.

Terry Brewer, CEO and president of Brewer Science, believes artificial intelligence is the necessary disruptor that will change the industry.

“We’ve been on a continuous improvement pathway in this industry for a very, very, very long time. We need a disruptor, and AI can serve as that disruptor.”

Advances in Computing Power Slowing

Transistor technology has improved tremendously over its lifespan, however, computing power appears to be slowing down. In the 40 years between 1970 and 2010, chips grew from supporting one thousand transistors to supporting one billion transistors. However, we’re seeing a leveling off of Moore’s Law that poses problems for next-gen technologies that are just beginning to gain momentum.

In order to address this negative trend, materials developers will need to re-evaluate fundamental approaches to semiconductor R&D. Sanjay Natarajan from Applied Materials stresses the importance of continuously unlocking new levels of computing efficiency.

“Computing power has significantly stalled in the past seven or eight years. That is really going to have severe consequences for the future of things like AI…the key to doing something useful is really to continue driving Moore’s Law…where you reliably get more efficient computing every year.”

Designing Hardware to Meet Future Needs

Computer processing will need to evolve in the future as we demand more and more of next-gen technologies. Training AI systems to carry out higher-order functionality will require hardware that can adequately support activities, such as pattern matching and statistical learning. Because of the amount of labeled data available today, machine learning models are more powerful than ever, but they come with heavy processing requirements.

AI R&D has important implications for the semiconductor and components space. Intel CTO, Michael Mayberry, acknowledges that physical hardware development is going to be influenced heavily by the capabilities that we want AI systems to possess. So how is he going to approach the next revolution in the semiconductor space?

“And as I learn more about AI, I learn the bounds of my ignorance more and more and more. And the original question was, ‘How are you going to solve all these problems?’ I admit I don’t know. But I’m starting to know what I don’t know, so therefore we can go and work on those pieces.”