Americas Technology: Semiconductors: EDA: Poised to begin capturing Agentic AI value with potential revenue acceleration—Buy CDNS, SNPS

研报英文原文证据摘录

Americas Technology: Semiconductors: EDA: Poised to begin capturing Agentic AI value with potential revenue acceleration—Buy CDNS, SNPS

Goldman Sachs Americas Technology: Semiconductors

Chip design labor shortage could drive a material EDA revenue opportunity through 2030

Bottom Line: The semiconductor industry has a long-standing labor problem, with

the world producing new chip designers at just 2%-3% a year—yielding ~255k

engineers by 2030—relative to the 325k design engineers we estimate the industry

will require to service surging AI-driven demand.

According to a 2022 industry workforce study by the Semiconductor Industry

Association (SIA) and Boston Consulting Group (BCG), as of 2021 the industry already

expected a shortfall of roughly 23,000 chip designers by 2030. Since that study (which

predates the 2023 generative AI boom), hyperscalers and AI labs have quickly built their

own chip programs, while merchant chipmakers have cut design cycles from two years to

one. Layering in that incremental demand implies a global shortage of ~70k designers by

2030 based on our estimates. As AI design tools are the only supply response available in

the short term, our estimates suggest they could augment the industry by ~48k

engineer-equivalents by 2030—enough to cover the shortage the industry had five

years ago, but not the demand the AI boom has since created. We believe Cadence and

Synopsys—the two market leaders in EDA tools—are best positioned to convert a

structural labor deficit into a new revenue stream—roughly ~$3.7bn of incremental,

consumption-priced AI software by 2030 (~$11bn in cumulative revenue in 2026-30),

on top of a ~$14bn base of core EDA revenues, none of which is reflected in Street

numbers today.

Exhibit 1: Global chip designer supply vs.