According to the chief technology officer of US chip-maker Nvidia, Michael Kagan, cryptocurrencies do not add any value to society. Despite selling a large number of powerful processors to the sector, Kagan said that the development of artificial intelligence chatbots, such as Nvidia’s ChatGPT, was a more worthwhile use of processing power than mining cryptocurrencies.
Nvidia never fully embraced the crypto community and, in 2021, the company released software that limited the ability to use its graphics cards to mine the popular Ethereum cryptocurrency. This was done to ensure that supply went to its preferred customers, which include AI researchers and gamers. Kagan justified the decision by explaining that the limited value of using processing power to mine cryptocurrencies made it less attractive than other uses.
Kagan believes that chatbots like ChatGPT are more useful for society because everyone can create their own programme, and it can be adapted to the user’s specific needs. Crypto, on the other hand, is similar to high-frequency trading, an industry that has led to a lot of business for Mellanox, a company that Kagan founded before it was acquired by Nvidia. He did not believe that crypto would do anything good for humanity and compared it to other industries that people engage in just because they are profitable.
Nvidia is known for producing powerful graphics cards for PC gamers, but the company’s products became central to the AI boom almost by chance. The computationally intensive work of training a new AI system works significantly faster on the types of simple yet powerful processors that had been adopted by gamers.
Recently, Microsoft purchased tens of thousands of Nvidia’s AI-focused processors, and Amazon and Oracle have also bought thousands of the company’s chips. Nvidia also rents access to the chips directly through its DGX cloud service, starting at just under $37,000 per month for eight H100s wired together in a “cluster”. Nvidia’s CEO, Jensen Huang, believes that the company’s technology will reinvent nearly every industry, describing it as the engine behind “the iPhone moment of AI”.
Last year, Nvidia’s attempted $40bn takeover of the UK-based tech firm Arm failed due to regulatory difficulties.