Nvidia has spent the last several years becoming the financial engine of the artificial-intelligence boom.

Now the chip giant is turning some of that enormous cash generation back toward its own shares.

Nvidia's board has authorized an additional $150 billion for its stock-repurchase program, creating the largest increase in buyback authorization in the history of the company and pushing its remaining repurchase capacity to $235 billion. Nvidia expects to execute the remaining program through fiscal 2028.

The size of the announcement is difficult to overstate.

Apple's record $110 billion authorization in 2024 had set the previous benchmark for annual buyback increases. Nvidia has now surpassed it by $40 billion. Reuters noted that Nvidia's additional authorization is larger than the entire market capitalization of about 84% of companies in the S&P 500.

Yet the announcement is not merely about returning money to shareholders.

It is also a message about how Nvidia sees its own future.

The company is effectively saying that the AI boom is generating enough cash to support three competing priorities at once:

Build more technology.

Invest in the AI ecosystem.

Return capital to shareholders.

CEO Jensen Huang said Nvidia's growth is being driven by what he described as a “once-in-a-generation” shift toward AI and accelerated computing, adding that the company's cash generation gives it capacity to invest while returning capital to shareholders.

That combination is what has attracted so much attention from investors.

Nvidia's financial position is the foundation of the buyback.

The company has become one of the biggest beneficiaries of the explosion in AI infrastructure spending.

Its processors power training and inference workloads across cloud providers, technology companies, AI laboratories and enterprise customers.

That demand has transformed Nvidia's balance sheet.

The company ended its July quarter with $22.44 billion in cash and cash equivalents, while revenue growth expectations remain extraordinarily high. Reuters reported that Nvidia expects around 70% revenue growth for fiscal 2028, despite growing concerns about how long the current AI infrastructure spending cycle can maintain its pace.

The key question is what the company does with that money.

Nvidia could simply accumulate cash.

It could increase dividends.

It could make acquisitions.

It could invest in startups.

Or it could buy back its own shares.

The company has chosen a combination of these strategies, but the latest authorization puts shareholder returns on an entirely different scale.

There is an important distinction that investors need to understand.

The $235 billion figure is an authorization, not a promise that Nvidia will immediately spend that amount.

The board has given management permission to repurchase shares up to that amount under the program's terms.

The company expects to execute it through fiscal 2028, meaning purchases can occur over an extended period and may vary depending on market conditions, cash flow and management decisions. Nvidia has not said that $235 billion will be spent at once.

That matters because buybacks operate over time.

If Nvidia repurchases shares when its stock price is lower, the same amount of money can retire more shares.

If it buys at a higher price, fewer shares are removed.

The effect on earnings per share therefore depends partly on execution price.

At the most basic level, a share repurchase reduces the number of outstanding shares.

If a company's profits remain unchanged while the share count falls, earnings are divided among fewer shares.

That can increase earnings per share.

For a company like Nvidia, where profit growth has been extraordinary, buybacks can therefore act as another tailwind for per-share financial metrics.

The effect is not guaranteed to make a stock rise.

A company's market value depends on much more than share count.

Revenue growth, profit margins, competition, valuation multiples, interest rates and investor expectations all remain important.

But buybacks can alter the supply-and-demand balance for shares.

A company repeatedly purchasing its own stock becomes a source of demand.

That is why analysts sometimes describe repurchase programs as a potential support for share prices.

Barron's previously reported that technology-company buybacks could help drive another leg of the broader sector rally as cash-rich companies increase shareholder returns.

The buyback announcement arrived after a period in which Nvidia's share performance had lagged some of its semiconductor peers.

Through Friday, September 25, Nvidia was up just over 20% for the year, according to Reuters, roughly in line with the Nasdaq 100 but well behind AMD and Intel. Nvidia shares rose as much as 3.6% on Monday following the buyback announcement before the gains moderated.

That makes the timing interesting.

The company is not announcing the huge buyback after an uninterrupted surge that has left valuation at extreme levels.

Instead, management is committing additional capital while the market is increasingly debating how durable AI's explosive growth will be.

Nvidia's forward earnings multiple has also fallen to roughly 16.5, its lowest level since January 2015, according to LSEG data cited by Reuters.

That lower multiple can be interpreted in different ways.

Some investors may see it as a reflection of a more attractive valuation after rapid earnings growth.

Others may view it as evidence that the market has become less willing to pay premium multiples as concerns about future growth increase.

The buyback does not resolve that debate.

It does, however, show that Nvidia's board is comfortable committing significant capital to its own shares.

The buyback becomes especially interesting because Nvidia is not behaving like a mature company with no investment opportunities.

The opposite is true.

The company is deeply involved in expanding the AI ecosystem.

Nvidia has invested in AI startups, cloud providers and infrastructure projects, and its broader strategy increasingly includes software and services alongside hardware.

The company also recently launched an Open Agent Safety Platform designed to help organizations secure increasingly autonomous AI agents. The platform combines Nvidia OpenShell and the Sentry reference system and launched with more than 100 partners, according to Nvidia.

That means Nvidia is still investing heavily in future growth.

The buyback is happening alongside those investments, not instead of them.

That is possible because cash generation has expanded so dramatically.

A share repurchase program only makes long-term financial sense if a company can continue generating substantial free cash flow after funding its operations and investments.

Nvidia's management is signaling confidence that it can do exactly that.

Huang has emphasized that Nvidia's cash generation gives the company the ability to invest in the technologies driving the AI transition while simultaneously returning capital to investors.

That statement matters because the biggest debate surrounding Nvidia is no longer whether AI demand exists.

It clearly does.

The harder question is how long the spending boom will last and whether the company can maintain extraordinary growth rates as the market becomes larger.

Competition is intensifying.

AMD is expanding its AI accelerator business.

Intel is trying to regain relevance in advanced computing.

Google has developed its own TPUs.

Amazon has its own AI chips.

Microsoft and Meta are also increasingly investing in custom silicon.

Customers are becoming more sophisticated too.

The largest technology companies are attempting to reduce their dependence on any single hardware supplier.

Nvidia therefore has to keep innovating.

The buyback gives Nvidia another financial lever.

If the company generates more cash than it needs for near-term investment, it can repurchase shares.

If AI demand accelerates and cash flow rises, the company can continue funding infrastructure and shareholder returns.

If growth slows, the buyback authorization gives management flexibility over how to deploy capital.

But authorization alone does not eliminate the underlying risks.

Nvidia still depends heavily on third-party manufacturers and suppliers.

The company itself warns that its future performance can be affected by technology changes, competition, manufacturing and packaging constraints, customer demand, regulatory developments and geopolitical conditions.

The AI market also remains unusually dependent on massive capital spending by a relatively small group of hyperscalers.

If those companies reduce spending, chip demand could eventually slow.

That would affect Nvidia's revenue and cash generation.

So the buyback is not an insurance policy against a weaker AI cycle.

It is a deployment decision based on current and expected cash-generation capacity.

One of the more controversial elements of Nvidia's strategy is that it has invested directly in companies that are building AI infrastructure and purchasing Nvidia hardware.

Reuters has noted that this has drawn scrutiny from some investors because Nvidia's investments can indirectly support demand for its own products.

That does not mean those investments are necessarily problematic.

It simply highlights how intertwined the AI ecosystem has become.

Nvidia is simultaneously:

A chip designer.

A software platform provider.

An investor.

An infrastructure partner.

And now one of the largest corporate buyers of its own stock.

That is a very different company from Nvidia before the generative-AI boom.

The math behind the buyback is simple even if the overall market impact is not.

Every share repurchased is removed from circulation.

Assuming profits continue to grow, fewer shares can mean greater earnings per share.

That can affect how investors calculate the company's valuation.

It can also make future earnings growth look stronger on a per-share basis even if total profits increase at a slower rate.

But investors should distinguish financial engineering from operating performance.

A buyback cannot create revenue.

It cannot make AI demand stronger.

It cannot prevent competitors from gaining market share.

It cannot lower the cost of electricity required to run data centers.

It simply changes the distribution of ownership and can improve per-share financial metrics.

Nvidia's underlying business performance remains the decisive factor.

The record authorization therefore creates several checkpoints.

Investors will want to see how quickly Nvidia actually executes the repurchases.

They will watch free cash flow.

They will monitor gross margins.

They will track AI accelerator demand.

And they will compare Nvidia's growth against increasingly aggressive competition.

The company has already made its capital-allocation preference clear.

It believes its future cash generation is strong enough to fund extraordinary investment and still return enormous amounts of money to shareholders.

For Nvidia, that confidence is reflected in a figure rarely seen in corporate finance:

$150 billion.

That one number has lifted the remaining buyback authorization to $235 billion and placed Nvidia in territory previously associated with the largest cash-generating companies in the world.

The AI boom made Nvidia a chip powerhouse.

Now it is giving the company enough financial firepower to become something else as well — a giant buyer of its own stock.

Whether that ultimately changes the long-term trajectory of the shares will depend on earnings, cash flow and the durability of AI demand.

But one thing is clear.

Nvidia has accumulated enough financial muscle to make its capital-allocation decisions a market event all by themselves.