Bank of America sees Nvidia very differently from investors worried that the AI boom has gotten ahead of itself. Analyst Vivek Arya has a $350 price target on the stock and says Nvidia could be trading at a 34% to 50% discount to its underlying value. With Nvidia shares trading around $220 on August 19, that target represents roughly 59% upside from today’s price.
But that does not make the path to $350 simple. Nvidia is no longer just selling the chips at the center of the AI buildout. It is also putting enormous amounts of capital behind the companies and infrastructure buying those chips. That can strengthen demand and deepen Nvidia’s reach across the industry. It also ties more of the company’s future to whether AI spending keeps producing the returns investors expect.
Why Bank of America Thinks Nvidia Is Cheap
Arya’s argument centers on free cash flow, Nvidia’s scale, and management’s decision to secure more of the AI supply chain itself. The company is not waiting for customers to solve every bottleneck. It is helping line up compute capacity, land, power, and financing where those resources can support future Nvidia systems. BofA sees that as a strategic advantage and believes investors are putting too much weight on the risks created by those relationships. The $350 target is aggressive, though. Nvidia’s 52-week high is $236.54, reached on May 14, meaning the target sits nearly 48% above that level. A lot still has to go right for shareholders to get there.
Nvidia Is Becoming More Than a Chip Supplier
The concern is not hard to understand. Nvidia still sits at the center of AI computing, but it is increasingly funding parts of the ecosystem surrounding those sales. OpenAI said in February that Nvidia was contributing $30 billion to its new funding round. Anthropic separately announced that Nvidia had committed to invest up to $10 billion in the company. Those relationships can help major customers expand faster and can strengthen Nvidia’s position across the industry. They also raise a fair question for shareholders: how much future demand is being supported by Nvidia’s own capital, and how much would exist at the same scale without it? That is not an accounting accusation. It is a financial risk worth understanding.
The $105 Billion Ohio Commitment Raises the Stakes
The newest example is Nvidia’s agreement tied to SB Energy’s PORTS-Pike data-center campus in Ohio, where OpenAI will be the tenant. An SEC filing says Nvidia’s aggregate payment obligation under its initial guarantees is capped at $105 billion. That is not the same as Nvidia writing a $105 billion check today. The obligations are subject to specific conditions and are expected to become effective as facilities are placed into service beginning in 2028. OpenAI must also reimburse Nvidia for amounts Nvidia actually pays under the guarantees. Even with those protections, the scale matters. Nvidia is taking a much larger role in making sure AI infrastructure gets built, financed, powered, and filled with its hardware.
Why Retirees and Long-Term Investors Should Care
For investors already living on their portfolios, the important question is less exciting than whether Nvidia eventually reaches $350. It is how much risk one stock is allowed to carry. A massive company can still face sharp price swings, execution problems, and changing demand. The Department of Energy says data-center growth is already contributing to rising electricity demand and a pressing need for more transmission infrastructure. If AI demand keeps expanding, Nvidia’s strategy could prove extremely valuable. If projects are delayed, financing becomes harder, or customers pull back, those same commitments could become a heavier burden. A bullish analyst target is useful context, but it should never be confused with certainty.