Global chipmakers to invest over $300m in AI-designed chips this year

The world’s leading semiconductor companies will likely spend more than $300 million in the development of artificial intelligence (AI) to help design semiconductors faster this year as a global chip shortage continues to put a strain on demand, Deloitte Korea said Friday in its report.

According to the report, global chipmakers such as Samsung Electronics, Intel and Qualcomm are expected to invest this amount in AI design tools for chips. By growing 20 percent annually for the next three years, the figure will likely surpass $500 million in 2026.

It may not be a large amount of money compared to the anticipated value of this year’s global chip market of $660 billion, but it is significant for the outsized return on investment, the consulting firm said.

According to Deloitte’s analysis, AI design tools enable chipmakers to save time and money on manufacturing semiconductors, alleviate the talent shortage and drag older chip designs into the modern era. The advanced tools can also increase supply chain security and help mitigate the next chip shortage, it said.

“The advanced AI tools are anticipated to alleviate a range of chip industry’s issues from intensifying foundry competition for micro-processing and increasing cost of chip manufacturing to persisting chip shortage,” said Choi Ho-kye, leader of technology, media and telecommunications at Deloitte Korea.

The report pointed out that the AI design tool developed by the Massachusetts Institute of Technology increased the energy efficiency to more than double that of human-designed circuits. Taiwan's MediaTek was able to cut its key professor component’s size and power consumption by 5 percent and 6 percent, respectively, using AI technology.

Since electronic design automation (EDA) vendors have made tools for chip design for decades, the size of the global EDA industry was estimated to stand at $10 billion last year, growing at about 8 percent annually. For the next five years, the growth in advanced AI tools for chip design is expected to be more than twice that of EDA tools and more than triple the growth rate of chip sales, it said.