The AI boom has brought strong cash flow.

On the evening of September 28th Beijing time, NVIDIA announced that the company's board of directors had approved an increase of the stock repurchase plan by 150 billion US dollars, raising the remaining total authorized repurchase amount to 235 billion US dollars. It is expected to be completed by the end of the fiscal year 2028. This is the largest single increase in stock repurchase authorization in NVIDIA's history, surpassing Apple's 110 billion US dollar repurchase plan announced in May 2024. This is also the second time in a short period that NVIDIA has significantly increased the repurchase amount - about four months ago, the company just added an additional 80 billion US dollars in stock repurchase authorization. After the announcement, NVIDIA's US stock price rose by more than 3% in the morning of Monday.

NVIDIA CEO Jensen Huang said that the company's growth benefited from the generational transformation of the computing platform to the AI and accelerated computing platform. The strong cash creation ability enables NVIDIA to continue investing in related technologies and also return capital to shareholders. This authorization reflects NVIDIA's confidence in future long-term opportunities. In March this year, Jensen Huang said in an interview that NVIDIA needed cash to support suppliers and partners and invest in the ecosystem, but still had a large amount of free cash flow, and stated that 50% of the free cash flow would be used to return to investors, a ratio higher than last year.

The previous financial report showed that NVIDIA's revenue in the second fiscal quarter of 2027 was 96.2 billion US dollars, an increase of 106% year-on-year and 18% quarter-on-quarter. It was the fourth consecutive quarter with a year-on-year growth rate acceleration, higher than the market expectation of approximately 92.1 billion US dollars. The adjusted earnings per share were 2.22 US dollars, also exceeding the market expectation of 2.09 US dollars. NVIDIA's management previously expected that the company's revenue in the fiscal year 2028 would increase by approximately 70%. As of the end of the second fiscal quarter, NVIDIA held 22.44 billion US dollars in cash and cash equivalents. Jensen Huang said in an interview that the current market is experiencing the largest infrastructure construction wave in human history, and NVIDIA is at the core of it. With further growth in cash flow in the coming years, the company hopes to continue returning more funds to shareholders.

At the same time as the rapid growth in performance, NVIDIA's valuation has significantly declined in recent years. According to data compiled by LSEG, NVIDIA's expected price-earnings ratio for the next 12 months based on future earnings is approximately 16.5 times, the lowest level since 2015 and significantly lower than the average of about 30 times over the past 15 years. Some market observers believe that in the context of a significant valuation decline, expanding the repurchase may also indicate that the management believes that NVIDIA's stocks have become more attractive relative to the long-term growth prospects.

In May this year, NVIDIA's market capitalization reached 5.67 trillion US dollars. In recent days, several industry practitioners in the AI field have publicly discussed AI security issues and potential hazards. Under the influence of market sentiment and other factors, the share prices of several semiconductor stocks in the US fell on September 14th. On that day, NVIDIA's market capitalization evaporated by 176.6 billion US dollars, but then rebounded. On the close of September 25th, NVIDIA's market capitalization was 5.42 trillion US dollars, still being the only stock in the US with a market capitalization exceeding 5 trillion US dollars.

Jensen Huang has also recently expressed his views on AI security issues. At an open event earlier this month, in response to the statement that "AI should slow down its development", Jensen Huang said that the related panic created a false choice between rapid innovation and AI security. Later, at an AI event in the UK, he said that AI security is crucial, and enterprises should continue to keep up with the pace and suspend the launch of products if they believe that certain products have safety hazards.

On September 28th Beijing time, NVIDIA also launched an open intelligent agent security platform. This platform can be used throughout the entire process from intelligent agent testing to deployment to enhance AI security. It covers the software running the intelligent agents, the hardware and computing layer supporting the operation of the agents, and the full-stack governance and control of the robot systems performing tasks in the physical world. NVIDIA stated that recent security incidents have highlighted the necessity for organizations to equip themselves with open and customizable tools to strengthen the control of long-running intelligent agents. More than 100 institutions, including Accenture, Cadence, Cisco, Dassault Systèmes, Deloitte, Perplexity, and Microsoft, have already adopted the platform technology. Huang Renxun believes that only by addressing AI security issues can the extraordinary potential brought by AI to society be fully unleashed.

In terms of AI ecosystem investment, NVIDIA currently directly invests in a large number of AI startups, cloud service providers, and model developers such as OpenAI. NVIDIA disclosed that its investment portfolio includes 13 listed companies and 229 non-listed companies, and the return on investment of the exited projects exceeds three times the amount invested. The company stated that in the future, it plans to return the excess free cash flow generated from these investments to shareholders through stock buybacks and gradually increasing dividends.

At the same time, NVIDIA's expanding AI financing layout has also begun to attract market attention. Some investors are concerned that after the company invests in cloud service providers and AI enterprises, these enterprises may use the financing to purchase NVIDIA chips, potentially forming a circular financing that indirectly drives their own demand. NVIDIA announced last month that it has collaborated with several Wall Street financial institutions to establish an AI computing power financing platform, planning to leverage over 500 billion US dollars of third-party capital for AI infrastructure construction, and also participating in some specific data center projects through credit support and other means.

In addition, NVIDIA has signed long-term data center lease agreements worth approximately 20 billion US dollars to support the expansion of the AI ecosystem and expects to transfer related lease agreements to other enterprises in the future. NVIDIA's Chief Financial Officer Colette Kress previously stated that the AI industry is expanding at an extremely fast pace, and fast-growing enterprises need more capital support. NVIDIA is necessary to help drive this industry's flywheel to continue operating. 

Disclaimer: The article is from the internet. If there are disputes, please contact customer service.