Nvidia has become so profitable from the artificial-intelligence boom that it now has a new problem familiar to only a tiny group of corporate giants:

What should it do with all the cash?

The company's answer arrived Monday in spectacular fashion.

Nvidia's board approved a $150 billion increase to its stock-repurchase authorization, bringing the total remaining capacity under its buyback program to approximately $235 billion. Nvidia said it expects to execute the program through fiscal 2028.

That is not merely another corporate buyback.

According to Nvidia, it is the largest increase in repurchase authorization in the company's history — and according to Reuters and other reporting, it represents the largest such increase on record. It also eclipses Apple's previous $110 billion authorization increase in 2024.

The announcement says something important about the financial transformation that artificial intelligence has created for Nvidia.

This is no longer simply a chip company benefiting from a new product cycle.

It is a cash-generation machine operating on a scale large enough that its capital-allocation decisions can move the market.

The buyback only makes sense because Nvidia can generate the cash to fund it.

The company had around $22.44 billion in cash and cash equivalents at the end of July, according to Reuters. Its balance sheet has expanded dramatically as AI accelerator sales have driven record revenue and profits.

The company reported quarterly revenue of roughly $96.2 billion, while free cash flow reached approximately $21 billion, according to Yahoo Finance's reporting on the buyback.

Those numbers explain why management can consider a buyback measured in hundreds of billions of dollars.

But there is an important distinction between cash already sitting on the balance sheet and future cash generation.

Nvidia's $235 billion authorization is not a check being mailed out immediately.

It is a board-approved ceiling on future repurchases.

The company expects to execute the program over several years.

That gives Nvidia considerable flexibility over timing.

When a company buys its own shares, the amount of stock it can retire depends on the price it pays.

Buybacks at lower prices retire more shares.

Buybacks at higher prices retire fewer shares.

That makes Nvidia's current valuation particularly relevant.

The stock closed Monday at approximately $228.86, according to Yahoo Finance historical market data, after rising more than 1.7% during the regular session.

The company's market capitalization is above $5.4 trillion, according to Reuters and Yahoo Finance reporting.

At that scale, even a $235 billion authorization represents only a fraction of the company's overall equity value.

Yahoo Finance calculated the authorization at roughly 4% of Nvidia's recent market capitalization.

That means the buyback is enormous in absolute dollars, but it will not suddenly remove a quarter or half of the company's outstanding shares.

Its effect will unfold gradually.

The basic mechanics are straightforward.

Imagine a company earns the same amount of profit next year but has fewer shares outstanding.

Those profits are divided among fewer shares.

Earnings per share rise.

If the company's profits are also growing at the same time, the impact can become more powerful.

Nvidia is in that unusual position.

Its revenue is still expanding rapidly because AI data-center demand remains enormous.

The company has projected roughly 70% revenue growth for fiscal 2028, according to Reuters and Bloomberg reporting.

That means the buyback is being layered on top of operating growth.

Investors therefore have two potential sources of per-share improvement:

The business generates more profit.

The company reduces the share count.

This is where the Nvidia story becomes more complicated.

A buyback can improve financial ratios.

It can signal management confidence.

It can provide demand for shares.

But it cannot create customers.

Nvidia still needs hyperscalers, AI developers and enterprises to keep buying its processors.

It still needs customers to spend billions of dollars building data centers.

It still needs to remain ahead of AMD, Google, Amazon and other competitors developing their own AI chips.

And it still needs the broader AI investment boom to continue.

The buyback is therefore not a substitute for growth.

It is an expression of confidence that the growth will produce enough cash to support both investment and shareholder returns.

Jensen Huang made that argument directly, saying Nvidia's cash generation allows the company to invest in technologies advancing AI while also returning capital to shareholders.

That is an important point.

Nvidia is not acting like a company that has run out of productive ways to invest.

It continues to spend heavily on its own technology.

It continues investing in startups.

It supports AI infrastructure projects.

And it is increasingly building software around its hardware.

On Monday, Nvidia also launched the Open Agent Safety Platform, a system designed to help organizations secure autonomous AI agents. The platform launched with more than 100 partners, according to Nvidia.

That means the company is simultaneously expanding its role in AI chips, software, infrastructure and security.

The buyback is occurring alongside that broader expansion.

This is another reason the $235 billion authorization matters.

Nvidia has increasingly used its enormous cash position to invest in the companies buying and building around AI technology.

Reuters has reported on Nvidia's investments in startups and cloud providers, noting that the company has developed a sizable investment portfolio as part of its broader strategy.

That creates an unusual feedback loop.

Nvidia sells chips.

Some customers use those chips to build AI infrastructure.

Nvidia invests in parts of that ecosystem.

Those investments can potentially generate financial returns while helping strengthen the overall AI market.

Critics may question how closely such investments are tied to demand for Nvidia's own products, but the strategy demonstrates how much the company's role has expanded beyond semiconductor design.

The timing of the buyback is particularly interesting because Nvidia's shares have not risen as much as those of some competitors in 2026.

Reuters noted that Nvidia's stock was up about 20% for the year as of late September, while AMD and Intel had delivered much larger gains.

That relative performance matters.

Nvidia remains the dominant name in AI chips.

But investors are no longer treating that dominance as sufficient by itself.

They want to know how much future growth is already reflected in the share price.

The company has been addressing that concern with a combination of higher revenue expectations and shareholder returns.

Corporate capital allocation communicates something.

A company that authorizes a giant buyback is effectively saying that management believes it can return capital without starving the business of funds needed for future growth.

Nvidia is making exactly that calculation.

Huang described the AI transition as a “once-in-a-generation platform shift” and said the new authorization reflects the company's confidence in the long-term opportunity.

But investors should distinguish a confidence signal from a guaranteed outcome.

Management can be confident.

Markets can disagree.

The future depends on demand, competition, technology and the economics of AI infrastructure.

The Nvidia buyback arrives at a moment when Wall Street is debating how sustainable the AI capital-spending boom really is.

AI companies are committing enormous sums to computing infrastructure.

Anthropic's latest IPO disclosures, for example, show more than $518 billion of long-term infrastructure commitments.

The broader AI ecosystem is increasingly relying on debt, cloud capacity, semiconductor purchases and power-intensive data centers.

That is excellent news for companies supplying the hardware.

But it also raises a longer-term question.

How much spending can the industry sustain before companies begin demanding stronger returns from their AI investments?

Nvidia's business model depends heavily on that spending remaining strong.

The $235 billion figure sits at the intersection of two very different Nvidia stories.

The first is the story of an extraordinary AI business generating enormous cash flow.

The second is the story of a stock whose valuation is now so large that even its own capital-allocation decisions become market-moving events.

The company has enough financial capacity to buy tens of billions of dollars of its own shares while simultaneously spending heavily on research, infrastructure and strategic investments.

Few companies in history have operated at that scale.

Investors will now watch how Nvidia uses the authorization.

A board approval is not the same as an immediate reduction in shares outstanding.

The actual impact will depend on the number of shares repurchased, average purchase price and pace of execution.

The company has until fiscal 2028 to work through the authorization.

That gives management significant flexibility.

Nvidia could accelerate repurchases during periods of weakness or spread purchases over several years.

The eventual effect on earnings per share will depend on that execution.

Every dollar spent buying stock is a dollar not spent on something else.

Nvidia could theoretically use additional capital for acquisitions.

It could invest more aggressively in software.

It could finance infrastructure partnerships.

It could invest in AI startups.

It could simply retain the cash.

The board has decided that returning a substantial amount of money to shareholders belongs high on the priority list.

That is a significant shift for a company whose explosive growth has historically encouraged investors to focus almost entirely on reinvestment.

Nvidia shares rose in early trading after the announcement, gaining around 1.7% in premarket trading and later finishing Monday around $228.86.

But the broader stock market fell that day.

The S&P 500 declined 0.8%.

The Nasdaq lost 0.9%.

The Dow fell 0.7%.

Rising Treasury yields and oil prices overshadowed much of the optimism around Nvidia.

That is an important reminder.

Even a historic corporate buyback does not operate in isolation.

Nvidia remains part of a much larger market.

If rates rise dramatically, expensive growth stocks can come under pressure.

If AI spending slows, semiconductor demand can weaken.

If geopolitics disrupts supply chains, the consequences can extend across the technology sector.

The buyback cannot eliminate those risks.

Over time, a sustained repurchase program can reduce the number of shares available in the market.

That can support earnings per share.

It can increase demand for stock.

And it can potentially reduce volatility associated with some forms of shareholder selling.

The effect is especially interesting in a company generating as much cash as Nvidia.

A business with weak cash flow cannot sustain huge buybacks indefinitely.

A business generating tens of billions in annual free cash flow has more room.

Nvidia is betting that it will remain firmly in the second category.

The Nvidia buyback may ultimately mark the beginning of a new phase for the AI boom.

Early in the cycle, Nvidia was the company desperately trying to supply a market that could not get enough chips.

Now it has become a company with so much cash that it can return enormous sums to its own shareholders while continuing to invest heavily in AI.

That is a very different financial position.

It suggests that the AI boom has created not just extraordinary demand for computing but extraordinary cash generation for the companies best positioned to supply it.

The $150 billion addition is therefore more than a number.

It is a statement that Nvidia believes its own future cash flows will be large enough to fund both the AI revolution and one of the biggest shareholder-return programs ever authorized.

Whether that confidence ultimately proves justified will depend on what happens next in the AI economy.

Can hyperscalers keep spending?

Can Nvidia maintain its technological lead?

Can margins remain high?

Can competitors narrow the gap?

Can customers generate enough revenue from AI to justify the infrastructure bill?

Those are the questions that will determine the long-term economics.

For now, Nvidia has made one thing unmistakably clear.

The company does not merely have the cash to participate in the AI race.

It has enough cash to become one of the biggest buyers of its own stock in financial-market history.

And that could make the next chapter of Nvidia's story just as interesting on Wall Street as it is inside the data center.