Photo from Nvidia
Only five years ago, Nvidia branded itself as a GPU (Graphical Processing Unit) company. That made sense. GPUs were its main line of business: Nvidia created GPUs for ordinary consumers, professional gamers, artisans, and more recently, cryptocurrency miners and Artificial Intelligence (AI) engineers. All the major announcements and products from Nvidia would be targeted towards gamers, who were the primary consumers of their higher end GPUs.
More recently however, Nvidia rebranded itself as an AI company, stating that “AI is fundamentally changing how we work and live … a shift of this magnitude is going to require enterprises to rethink and re-architect their infrastructures”. Their newest lineup of gaming GPU cards have been criticized for a lack of innovation yet being even more expensive than the previous generation. Major technology news sources such as Gamers Nexus have stated that Nvidia has essentially neglected their gaming lineup in favor of AI. So what caused Nvidia to change its identity from a GPU company to an AI company?
The primary reason for Nvidia’s rebranding is due to the surge in Generative AI. Generative AI, which includes programs such as ChatGPT, which are based on deep learning architecture called transformers. These transformer-based models require large amounts of computational power to be trained on, especially GPUs that can handle such large inputs of data. Currently, there are only a handful of companies that manufacture GPUs capable of training large AI models: AMD, a Taiwanese company; Intel, which has only started to produce GPUs in the late 2019s, and finally Nvidia, which is currently the leading manufacturer of GPUs. In fact, Nvidia A100 GPUs, which are the most common GPU to train large models such as the GPT series, are only ever competed by the RTX 4090, Nvidia’s own consumer based GPU. Essentially, Nvidia currently has a near monopoly on a rapidly growing and demanding field. Such has led to Nvidia’s shift towards AI technology.
Cryptocurrency is also a major factor. The surge in cryptocurrency prices in 2021 to 2022, with bitcoin prices reaching near 60,000 dollars, has led to a significant increase in demand for GPUs also. GPUs are essential to “mine” cryptocurrency, as cryptocurrency is produced by computers solving complex equations to earn cryptocurrency as a reward, which requires large amounts of computing power—something that GPUs are specifically manufactured to do. Nvidia saw prices skyrocket during this time, with cryptocurrency miners scavenging for GPUs, offering to pay 2 to 3 times the market price. For Nvidia, Professional gamers and artists went from their most lucrative consumers to 3rd lucrative.
Proponents of Nvidia might argue that the company has no responsibility to continue providing reasonably priced GPUs for ordinary consumer use. Nevertheless, Nvidia must continue innovating for their own good. When considering the entire GPU market, rather than focusing on AI, Nvidia is not without competition–AMD has already established contracts to provide GPUs for Sony’s Playstation and Microsoft’s Xbox, two of the largest companies in the console gaming industry. Even when considering GPUs for AI, Google’s Tensor Processing Unit (TPU) , specifically designed to train large language models, is continuing to increase its prominence in the AI industry as a replacement to Nvidia’s GPUs.
Nvidia undoubtedly has larger concerns for cryptocurrency and AI rather than its normal consumers. While Nvidia has every right to pursue profit in these areas, they should also take care to not forget that in the end, advancements for the common consumer such as gamers and designers are going to be the main factor in integrating artificial intelligence in our lives.
