3.) Artificial intelligence
|(Image source: silicium ©P.JAYET/CEA)|
Artificial intelligence (AI) is on everyone’s list of trends. And rightly so as IDC estimates that the AI technology market should grow from $8 billion in 2016 to over $47 billion in the year 2020.
AI tech is being used in facial recognition, cybersecurity, and edge and cloud computing, Another trend, specialized, AI-enabled chips, gained increased attention this year in the semiconductor and EDA chip design tool industries when an entire day was dedicated to AI at SemiconWest, one of the largest semiconductor conferences around.
As keynote speaker for AI Design Forum which kicked-off the AI day event, Synopsys chairman and co-CEO, Aart de Geus, highlighted the shift from the Moore’s Law era of computational chips to the newer chip architectures required by AI. He sees simulation, modeling, and virtual prototyping as crucial areas affected by this change.
From a business perspective, AI and machine learning continued to dominate investment in new semiconductor chips in 2019. According to Wally Rhines, CEO emeritus of Mentor, a Siemens business, venture capital funding of 30 fabless AI companies in 2018 totalled $2.3 billion. In 2019, year-to-date funding has exceeded $850 million so far. While lower than 2018 at this time, the funding is more diverse, covering a broader base of companies and applications.
“Machine learning also had a major impact on EDA, with dozens of new and improved design tool capabilities making their appearance by applying AI techniques to traditional problems like simulation, optical proximity correction, timing closure, place/route, and yield enhancement,” explained Rhines.
One example of an emerging AI application was in edge computing. In 2019, two of Europe’s leading R&D centers, CEA-Leti of France and the Microelectronics Institute within Germany’s Fraunhofer Society, announced joint research the development of neuromorphic computing techniques for edge computing. This computing approach looks to create electronic circuits that mimic biological architectures present in the human nervous system.
The institutes’ work toward edge-AI systems is building up Leti’s strength in fully-depleted silicon-on-insulator (FD-SOI) chip design and the expertise of both Fraunhofer and Leti in 3D packaging. There is also the likelihood that it will draw upon finFET architectural research by another EU R&D powerhouse, Belgium’s Imec.