The profound impact of artificial intelligence is rapidly moving beyond the digital realm and into the physical world, creating massive new market opportunities that demand the specialized compute power that Nvidia provides. This was the central thesis delivered by Denny Fish, Portfolio Manager at Janus Henderson, during a discussion on Bloomberg Tech regarding Nvidia’s strategic positioning, anchored by the headline news of its $1 billion investment in an AI drug discovery lab with pharmaceutical giant Eli Lilly. Fish spoke with the Bloomberg anchors about the accelerating momentum of AI deployment and the foundational role Nvidia plays, suggesting that the transformation of physical industries like healthcare and manufacturing represents an opportunity as large, if not larger, than the current boom in large language models.

The discussion immediately highlighted the significance of the Eli Lilly partnership. For many investors, the AI revolution has been defined by the "digital manifestation" of technologies like ChatGPT. Fish acknowledged this focus, noting that generative AI has "grabbed all the attention over the last three years." However, he stressed the long-term economic and societal impact of applying AI to tangible problems. "The physical manifestation of AI is as big if not a bigger opportunity," he stated, pointing to drug discovery, robotics, and autonomous driving as key verticals. This strategic co-investment with a market leader like Eli Lilly is a calculated move by Nvidia to cement its hardware and software platform (CUDA) into the complex, high-stakes environment of pharmaceutical research.

Drug discovery stands out as the ultimate prize in the physical application of AI. The traditional process is notoriously slow, expensive, and prone to failure. By deploying massive computational power—the kind Nvidia specializes in—the time horizon for identifying and validating promising compounds can be drastically reduced. Fish elaborated on the potential payoff: "The real big prize out there is drug discovery. It’s finding solutions to problems that we just couldn’t get there without AI or would have taken much, much longer." For Nvidia, these partnerships are not merely about selling GPUs; they are about embedding their ecosystem deeply into the operational DNA of multi-trillion-dollar industries, ensuring long-term revenue streams built on mission-critical performance.

Beyond specific vertical applications, the infrastructure supporting the entire AI wave continues to dominate the investment landscape. Fish affirmed that the investments into AI infrastructure "are going to continue to be quite robust." This stability is crucial for founders and VCs looking at the long game. The key drivers are the hyperscalers—Meta, Google, Microsoft—who possess the free cash flow necessary to fund the immense data center buildout required to support both training and inference workloads. The sheer scale of this capital deployment—with Moody’s forecasting global data center related investments of at least $3 trillion over the next five years—underscores the foundational nature of the current tech cycle.

The conversation naturally moved from the training phase of AI (where large models are built) to the deployment phase, known as inference. Inference is where the rubber meets the road—where trained models are used daily by millions of users or embedded into physical systems like autonomous vehicles and industrial robots. Fish emphasized that inference represents the next major handoff in the AI curve. With new chips like Nvidia’s recently announced Rubin platform entering production, this focus on real-time, high-efficiency computation is gaining traction. "Inference is the next big handoff," he explained, "and so if you’re a believer that we are now in this period where we’re going to start seeing increasing deployment of these applications, inference is going to go through the roof." This deployment phase sustains demand far beyond the initial infrastructure installation, ensuring a prolonged and healthy market for advanced compute components.

When addressing concerns about the speed and cost of this buildout, particularly the massive capital expenditure required for data centers, Fish offered a nuanced perspective contrasting the current environment with previous speculative booms. He pointed out that while the numbers are staggering, the deployment of capital is "going to be pretty measured." This measured approach is necessitated by real-world constraints: the complexity of building 10-gigawatt data centers, securing labor, and ensuring adequate power supply. "This is hard," Fish noted, referencing challenges in places like Abilene, Texas. Unlike the dot-com era, where capital flowed quickly into easily deployed telecom infrastructure, the physical realities of AI deployment act as a natural governor, preventing unsustainable oversupply. This supply constraint, coupled with high demand, remains strategically advantageous for suppliers like Nvidia. As Fish concluded, when demand outstrips supply, it is "really good from a pricing standpoint, particularly when you’re the only game in town," a position Nvidia currently enjoys in several crucial aspects of the AI infrastructure stack.

Nvidia and Eli Lilly The Physical Manifestation of AI

AI Daily Digest

Nvidia and Eli Lilly The Physical Manifestation of AI

AI Daily Digest