Article: The Gold Rush of Artificial Intelligence Suppliers

Photo Source: Flickr

Artificial Intelligence (AI) and machine learning (ML) command attention today like never before. The market for AI-driven products continues to explode. For example, the company Nvidia (NVDA), which makes the microchips fueling the explosion in artificial intelligence, just exceeded a market capitalization of $1.17 trillion (as of 8/24/23). This previously relatively little-known company making chips for AI has a greater valuation than the combined value of Tesla, General Motors, Mercedes Benz, Ford, and Honda, which combine to total $948 billion. The market sees tremendous value in the future of AI.

Such an expansion of interest in AI and its projected value has created a new gold rush, and the California Gold Rush of 1849 taught us the lesson that those who sold supplies such as boots, shovels, food, and blankets to the thousands of prospectors made more money than most of those who were mining for gold. Notably, the entrepreneur Samuel Brennan sold $5,000 a day worth of equipment (the equivalent of $120,000 in today’s money) to prospectors in the California Gold Rush. He built a store near Sutter’s Mill to cater to incoming miners, and for a time, that earned him the title of “Richest Man in California.” (pbs.org) AI such as chatGPT, developed by OpenAI, requires massive amounts of computation. Large language models such as chatGPT and Google’s Bard learn from text by playing a fill-in-the-blank game. By scanning through all the text on the internet, the models learn which word most likely follows other words. As the model grows, it will randomly remove words and guess which one should be there. When guessed correctly, the model gets reinforced to make the exact guess again, building the model’s strength. This model training takes billions of words and trillions of calculations, making the process expensive and time-consuming. Training costs for LLMs run is estimate to run about $5-10 million. (scineceblog.com) Therefore, the companies that make computer chips and hardware for AI training are supplying the AI gold rush of today. Nvidia stands as the current leader in the development of AI chips today, which helps explain its over $1 trillion valuation in the stock market. The company continues to roll out increasingly more powerful chip sets on which many companies building AI models rely today. However, many of the LLMs run on the cloud, which means that all the computing takes place on computers outside a smartphone or laptop, which can result in slow responses due to reliance on connection through the network. The venerable chipmaker Qualcomm (QCOM), headquartered in San Diego, CA, has an AI chip called Snapdragon that performs AI calculations inside a smartphone whiel not relying on web connectivity. The advantage in speed and not competing in the cloud opens the door to a different type of AI chip market althogether. Startups and established companies have announced new chipsets. For example, Advance Micro Devices, founded in Silicon Valley in the late ‘60s released the MI300x that they claim will challenge Nvidia in price and speed. (cnbc.com) The world of artificial intelligence has burst onto the scene in almost every aspect of industry, business, and society. Such a gold rush to make the most of this new technology invites many people inside and outside computer science to stake a claim in this profitable and expanding frontier. History has shown that some of the most tremendous profits from a gold rush accrue to those who supply the prospectors, not the miners. The spectacular rise of Nvidia to the sixth highest valued company in the world on the back of its industry-leading chips for building AI reaffirms the value of supplying the miner, not necessarily doing the mining. Many other chip makers such as Qualcomm and Advanced Micro Devices seek to make advances in AI chips to try and take a chunk of Nvidia’s market dominance and grab some of that AI gold.

Dr. Smith’s career in scientific and information research spans the areas of bioinformatics, artificial intelligence, toxicology, and chemistry. He has published a number of peer-reviewed scientific papers. He has worked over the past seventeen years developing advanced analytics, machine learning, and knowledge management tools to enable research and support high-level decision making. Tim completed his Ph.D. in Toxicology at Cornell University and a Bachelor of Science in chemistry from the University of Washington.

You can buy his book on Amazon in paperback and in kindle format here.