Nvidia (NVDA) has transformed from a video game chip maker into a dominant force in artificial intelligence (AI) chips over recent years, driving massive revenue growth—annual revenue increased by 2, 500% over the past decade. By entering the AI chip market early and continuously innovating, Nvidia secured and maintains market leadership. However, some investors worry about rising competition from other chip designers like Advanced Micro Devices and Broadcom, as well as from Nvidia’s own customers such as Amazon and Alphabet, which have developed their own AI chips for cost-conscious clients. Initially, Nvidia’s GPUs powered the training of large language models (LLMs), enabling them to learn and acquire knowledge.

Today, these GPUs also support the inference process, where LLMs “think” and “reason. ” Nvidia is betting on inference as a major growth area for AI chips. Its latest Blackwell GPUs outperformed previous models, delivering 10 times higher performance per watt and 10 times lower cost per token in MLCommons tests compared to the prior Hopper chips—highlighting significant advances in speed and cost efficiency. Nvidia CEO Jensen Huang emphasized the company's enduring edge during a recent earnings call, stating that “it's gonna take a long time before somebody is able to take that on, ” and affirming Nvidia’s leadership as “surely multiyear. ” While competitors may find niches—since some customers might not need the most powerful GPUs or may prefer customized chips—Nvidia remains the leading provider of the highest-performing AI compute power. Its strong gross margins above 70%, record revenues surpassing $130 billion in the latest fiscal year, and continued innovation position it well for future growth, particularly as inference workloads grow in importance for powering AI applications.