Nvidia's board of directors has authorized a one hundred fifty billion dollar increase to the company's share repurchase program, raising the remaining total program to two hundred thirty-five billion dollars. The authorization was announced Monday morning and circulated first through market-monitoring accounts, including @FirstSquawk, which posted shortly after eleven o'clock Eastern that "NVIDIA INCREASES SHARE BUYBACK APPROVAL BY $150 BILLION."
The buyback increase arrives alongside a separate Nvidia announcement: a new suite of software tools designed to keep artificial intelligence agents under control. According to Reuters, the company rolled out the tools from San Francisco on Monday, asserting that the technology would have blocked the cyberattack on Hugging Face, the AI coding platform. That claim was made by the company itself and has not been independently confirmed.
The two announcements together give the market a sizable pair of signals from the chipmaker — one financial, one operational — on the same morning. The buyback expansion is a balance-sheet decision that returns capital to shareholders and reduces the share count over time, while the security suite addresses a growing concern among enterprise customers: what happens when autonomous AI systems run without adequate guardrails.
What is not yet known is how the market will treat the buyback increase relative to the company's other priorities. Nvidia has not said how quickly it intends to execute the repurchase authorization, nor whether the two hundred thirty-five billion dollar remaining program will be deployed uniformly or opportunistically. It is also not yet known whether the AI agent security suite will be bundled with existing Nvidia enterprise offerings or sold as a separate product line.
The Hugging Face reference in Nvidia's announcement is notable. Hugging Face is a widely used platform for hosting and sharing AI models, and an attack on its coding infrastructure would represent a significant security event in the AI developer ecosystem. Nvidia's assertion that its new tooling would have blocked that attack is a claim made by the company in a press context, not an independently verified finding.
Next News Network could not independently verify the details of the Hugging Face cyberattack or Nvidia's assertion that its software would have prevented it. The buyback authorization figure, by contrast, is a corporate disclosure that appears in Nvidia's own board-approved announcement.
Market participants have been watching Nvidia closely through a year of extraordinary attention to AI infrastructure spending. The company's share price has been volatile at times, and a buyback authorization of this size is typically read as a signal that management views the stock as undervalued or that the company has more cash than it needs for operations and research. Nvidia has not said which of those rationales drove the decision.
The security suite announcement adds a second dimension. Enterprises deploying AI agents — autonomous systems that can take actions on behalf of users — have raised concerns about what happens when those agents encounter malicious code, attempt unauthorized actions, or are manipulated through prompt injection. Nvidia's tooling is aimed at that problem set. Whether it works as described is a question that will be answered by customers and independent researchers over time.
Together, the two announcements place Nvidia at the intersection of two live debates: how much capital should public companies return to shareholders, and how much responsibility AI platform vendors bear for the security of the systems built on top of their technology. Nvidia has taken a position on both.
The company's board authorization was described in its own announcement as raising the remaining total program to two hundred thirty-five billion dollars. That figure represents the total remaining authorization, not a single-quarter outlay. Actual repurchases depend on market conditions, cash flow, and management discretion.
It is not yet known whether Nvidia will disclose a timeline for the repurchases or whether the company will discuss the security suite's performance metrics publicly. Customers evaluating the tooling will likely want proof rather than assertions, particularly given the stakes involved in securing AI agent deployments.
The Hugging Face attack, if confirmed in detail, would be one of the more significant security incidents in the AI developer ecosystem to date. Nvidia's claim that its software would have blocked it is a marketing assertion until demonstrated otherwise. The company has not released technical details of how the tooling would have detected or prevented the specific attack.
For now, the buyback increase is the more concrete of the two announcements. It is a board-level authorization with a stated dollar figure, and it changes the company's capital return profile. The security suite is a product launch with a claim attached — that it would have stopped an attack that already happened. The first is a fact about corporate finance. The second is a claim about capability that remains to be tested.
Our Take
Nvidia is doing two things at once: rewarding shareholders with a massive buyback authorization and positioning itself as the company that can keep AI agents from going rogue. Both are smart moves on paper. The buyback signals confidence, and the security suite signals that Nvidia understands the anxiety enterprises feel about autonomous systems. But the security claim is where the company opens itself up. Saying your software would have blocked a specific attack that already happened is a bold assertion, and bold assertions in cybersecurity have a way of being stress-tested by people who do not care about your marketing calendar. If the tooling works, Nvidia has a genuine differentiator. If it does not, the company will have handed its critics a talking point. Either way, the buyback is the part that shows up in earnings reports. The security claim is the part that shows up in headlines when something goes wrong. Nvidia would do well to let the product speak for itself before making promises about attacks it did not see coming.
