Nvidia CEO Jensen Huang says AI won’t be the job killer everyone fears. Here’s why
AI will lead to more jobs, not fewer, Nvidia CEO Jensen Huang said Wednesday. In an interview with BlackRock CEO Larry Fink at the World Economic Forum in Davos, Switzerland, Jensen methodically explained how AI is different from other technological innovations of the past and how it can be an incredible engine for economic growth and prosperity. Jensen’s meeting with Fink, who co-chairs the WEF, was a high-level conversation about what artificial intelligence is, what it is not, what investments are needed, why and how much. At the heart of the change, the CEO of Nvidia Club said that artificial intelligence is a platform change. Examples of platforms include personal businesses, the internet, smartphones, and the cloud; these are the infrastructure, both physical and digital, on which the applications we all use in our daily lives are built. Just as these platforms laid the foundation for economic growth, so will artificial intelligence. While ChatGPT and other large language models are technically applications, the difference is that new, AI-focused applications will be built on top of these LLMs. Jensen sees five layers in building the AI platform: (1) enough energy to run everything; (2) chip/compute infrastructure dominated by Nvidia; (3) cloud infrastructure/services; (4) Artificial Intelligence models; and (5) the application layer, which is the foundational layer where the economic benefits of AI will be realized. Allaying not-so-reasonable fears of AI replacing human workers, Jensen offered a more optimistic perspective. Apart from the jobs created by physically creating AI infrastructure, such as electricians, plumbers, construction workers, Jensen said the implementation of AI will lead to greater demand for goods and services, which will lead to more hiring. Many workers may not be needed per customer, but customer growth that comes with a more efficient company can still result in demand for more workers. For example, Jensen said the number of radiologists is increasing because hospitals can now process more patients thanks to artificial intelligence. While AI can read scans, humans are still needed to review and confirm diagnoses. More patients, more scans to examine, and more radiologists were needed. The advantage is that by having AI perform the scans, radiologists have more time to spend with patients or discuss the results with peer clinicians. Jevons Paradox? As radiologists become better and more efficient at their job, it becomes easier to see a radiologist, resulting in an overall increase in demand from patients wanting to see radiologists. Radiology was a timely analogy, as Bristol Myers Squibb, the club’s name, said Tuesday it would work with Microsoft’s AI-powered radiology platform to develop and launch imaging algorithms aimed at accelerating early detection of lung cancer. These new tools, which can be used to analyze X-ray and CT scans, will help clinicians see hard-to-detect lung nodules and identify patients in the early stages of the disease. The alliance is the ultimate point where healthcare and AI rapidly intersect. Jensen, who was on stage with Fink on Wednesday, noted that there is a similar dynamic in nursing to what happens in radiology. Nurses spend about half of their time creating charts. But thanks to artificial intelligence, Jensen said nurses can spend more time visiting patients and provide the human touch that is often missing when a hospital is short-staffed. By reducing the nursing bottleneck, hospitals can see more patients faster. As a result, increased demand for hospitals leads to improved financial situations and therefore the opportunity to hire more nurses. According to Jensen, the key to understanding how AI will impact a particular job is to first distinguish between purpose and task. Going back to the radiology example, he said the goal of a radiologist is to diagnose patients, and at the end of the day, the job of both radiologists and nurses is to take care of people. Reading charts and scans is crucial for this purpose, but it is a task that must be done to achieve the ultimate goal of the job. By automating certain tasks, we can help people become better at their real jobs. Jensen argued that in the context of medical settings, AI should not result in fewer doctors and nurses, but should lead to more doctors and nurses spending more time with their patients and providing better outcomes. When asked about investments in artificial intelligence, Jensen pretty much scrapped the idea of a spending bubble. Instead, when asked whether investment levels were too low, his spending provided an easy way to consider whether a bubble was in the cards or whether more spending was necessary. He said it comes down to the fact that despite the millions of Nvidia GPUs (graphics processing units) currently available in the cloud, the GPU rental spot price is rising not just for the latest generation, but even for two generations of chips in the rearview mirror. He attributed this to the number of AI companies being created. When you see an increase in the rental price of a two-generation old GPU, it’s clear that supply (in general, not just with the latest chips) is falling well behind demand. Bottom Line While there are bubble-like areas in the market, the word has no place in the conversation when discussing names like Nvidia or mega-capacity cloud infrastructure players that generate real returns and trade at reasonable valuations. We’re stuck in the early stages of building AI infrastructure before we really start thinking about the application layer of the stack that touches every aspect of our daily lives. (Jim Cramer’s Charitable Trust is long NVDA, BMY, BLK. See here for a full list of stocks.) When you subscribe to the CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. Jim waits 45 minutes after sending a trading alert before buying or selling a stock in his charitable foundation’s portfolio. If Jim talked about a stock on CNBC TV, he would wait 72 hours after issuing the trading alert before executing the trade. THE ABOVE INVESTMENT CLUB INFORMATION IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY, TOGETHER WITH THE DISCLAIMERS. NO CIVIL OBLIGATIONS OR DUTIES EXIST OR SHALL BE RESULTING FROM YOUR RECEIVING ANY INFORMATION PROVIDED IN CONNECTION WITH THE INVESTMENT CLUB. NO SPECIFIC RESULT OR PROFIT CAN BE GUARANTEED.
