“This 40 percent achievement will eventually industrialize itself into AI workflows that require higher bandwidth and higher efficiency,” Gambetta said.
Roadmap to sub-1 nanometer nodes
As a company that conducts chip technology research, IBM does not manufacture commercial chips that might end up in AI data centers or consumer devices. Instead, IBM has partnered with semiconductor companies such as Rapidus in Japan to mass-produce its previous generation 2-nanometer node chips based on nanosheet architecture, or commercialized related technology in another partnership with Samsung in South Korea.
Other companies have followed IBM’s pioneering work without any direct collaboration. For example, Taiwan’s TSMC independently developed nanosheet transistors for its proprietary 2-nanometer node technology.
“Nanosheets have become the foundation of the next generation of transistor scaling,” said Huiming Bu, vice president of IBM Semiconductors Global R&D and IBM Research, during a media briefing. “Today, NanoSheet is adopted by all major foundries for most 3-nanometer chips and all 2-nanometer chips.”
IBM declined to name specific companies with which it may partner to commercialize the latest sub-1-nanometer node technology. But Boo expects production of commercial chips built at sub-1-nanometer nodes and incorporating the latest nanostack architectures could begin within the next five years, and possibly within a decade.
“This will replace nanosheets as the mainstream in leading foundries today, whether it’s CPUs or GPUs,” Bu said. “Within a decade, this will become another mainstream thing that we invented and help change the industry.”
<a href