Photo by Steve Johnson on Unsplash

Introduction

A recent PwC study shows that some experts believe AI may contribute up to $15.7 trillion to the global economy by 2030 but you probably aren’t invested in many of the best AI companies and this article should help rectify that situation. There are hundreds of such companies currently trading on North American stock exchanges but rather than attempting to select the one company that will be most successful we have created a portfolio - our model AI Semiconductor Stocks Portfolio - of 26 of the largest AI hardware (semiconductor) companies that have the clearest potential to continue to harness macro growth trends, and divided them into 6 sub-sectors of focus (see here).

Our Model AI Semiconductor Stocks Portfolio Constituents

The first 4 of the 6 sub-portfolios consists of those categories that account for the 4 specific steps in the manufacturing of semiconductors, namely:

1. the design of complex semiconductor chips using sophisticated electronic design automation (EDA) software, i.e. Our Model Pure-Play EDA Software Chip Design Stocks Portfolio, and consist of:
   • Cadence Design Systems (CDNS) which concentrates on system-level integration, packaging, and simulation-heavy workflows; and 
   • Synopsys (SNPS), a full-stack chip design, IP licensing, and software security provider.
2. the supply of equipment and material necessary to build and operate a semiconductor foundry, i.e., Our Model  Pure-Play Semiconductor Supplier Stocks Portfolio, and consists of companies that provide essential equipment for manufacturing chips, such as:
   • ASML Holding (ASML) that supplies Lithography machines used to transfer intricate patterns onto a silicon wafer;
   • Lam Research (LRCX) that focuses on the Etching process of transferring patterns on to the wafer surface, creating the desired circuit layout;
   • Applied Materials (AMAT) that specializes in Deposition which is a process of layering materials on to a wafer surface to impart electrical characteristics; and
   • KLAC Corp. (KLAC) that undertakes Metrology which refers to the science of measurement that ensures that each stage of chip fabrication meets strict quality and precision standards including measuring film thickness, detecting defects, verifying lithography accuracy, and ensuring process control at atomic and sub-nanometer levels.
3. the foundries that fabricate the chips based on the designs provided by other semiconductor companies, i.e. Our Model Pure-Play Semiconductor Foundry Stocks Portfolio; and consists of the following pure-play constituents: 
   • Taiwan Semiconductor (TSM) that specializes in fabricating leading-edge (3nm, 5nm, 7nm) nodes as well as mature nodes;
   • GlobalFoundries (GFS) and United Microelectronics (UMC) that focus on fabricating mature nodes (12nm–90nm and 28nm-90nm, respectively); and
   • Tower Semiconductor (TSEM) that focuses on analog, RF, power, and mixed-signal nodes for the automotive, industrial, and medical markets.
4. the companies that assemble chips into finished semiconductor components, test for defects, and provide the specialized packaging  necessary for shipping the chips around the world, i.e., Our Model Pure-Play Outsourced Semiconductor Assembly & Test Stocks Portfolio, and consists of:
   • Amkor Technology (AMKR) which benefits from reshoring and CHIPS Act incentives; and 
   • ASE Technology (ASX) which is more diversified, with EMS capabilities and a stronger presence in consumer electronics and networking.

The other two categories are those major companies that:

1. design and sell hardware devices and semiconductor chips themselves while outsourcing their fabrication (i.e., are fabless) to a specialized manufacturer; i.e., Our Model Pure-Play Fabless Semiconductor Stocks Portfolio; and consists of:
   •  Advanced Micro Devices (AMD); Broadcom (AVGO); Marvell Technology (MRVL); Monolithic Power Systems (MPWR); Nvidia (NVDA); and Qualcomm Incorporated (QCOM). All six companies are deeply embedded in the AI hardware stack - from GPUs and NPUs to networking ASICs and power management ICs.
2. do everything themselves in-house to control the entire production process from beginning to end i.e., Our  Model  Integrated Device Manufacturer Stocks Portfolio, namely:
   • Analog Devices (ADI) which is focused on EV battery management systems and industrial automation; 
   • Infineon Technologies (IFNNY); NXP Semiconductors (NXPI), and STMicroelectronics (STM) which are key suppliers for ADAS, powertrain electrification, and vehicle connectivity, and EV penetration;
   • Microchip Technology (MCHP) and Texas Instruments  (TXN) that have seen renewed demand in factory automation, smart grid, and medical devices; 
   • Micron Technology (MU) which is a major supplier of High Band Memory (HBM) and fifth-generation double data rate synchronous dynamic random-access (DDR5) memory in AI workloads; and
   • Intel's (INTC) that is undergoing both a foundry and AI accelerator push.

Conclusion

We believe that investing broadly in our selected 26 companies has a higher probability of producing positive results than attempting to pick a needle out of a haystack and over the next several months we’ll provide further detail and insight into both this sector and the constituents in it.

More By This Author:

What Bubble? All 6 AI Semiconductor Stock Categories Were Up Last Week
Latest AI News From Intel, Synopsys, Marvell, Nvidia, Cloudflare, Salesforce, ServiceNow And More
AI News From Meta, Amazon, CrowdStrike, Applied Materials And More