An announcement from the company brought bullish news, igniting the hot concept stocks of artificial intelligence (AI) such as Supermicro.

On Monday, October 7th, during the pre-market trading in the Eastern United States, Supermicro, which produces servers for AI training models, opened high and continued to rise. By the end of the pre-market session, it surged over 17.6%, marking the largest intraday increase since March 4th. It maintained an increase of over 10% and ultimately closed with an increase of around 15.8%, reaching $47.74. This set a new closing high since August 27th and the largest closing gain since May 15th.

Comments generally believe that the sharp rise in Supermicro's stock price is attributed to the strong demand for AI servers highlighted by the shipping data disclosed by the company that Monday.

In the announcement, Supermicro announced the launch of a complete liquid cooling solution, including powerful cooling liquid distribution units (CDU), cold plates, cooling liquid distribution manifolds (CDM), cooling towers, and end-to-end management software. The company stated that this solution can reduce the hardware costs of data centers, as well as the costs of cooling infrastructure that usually need to run continuously. It was mentioned that since June of this year, over 2000 liquid-cooled racks have been delivered, with shipments of more than 0.1 million GPUs equipped with this solution per quarter. The announcement stated:

"Supermicro recently deployed over 0.1 million GPUs with liquid cooling solutions (DLC) for some of the largest AI factories and other cloud computing service providers (CSP). As the power required for AI and High-Performance Computing (HPC) workloads per server approaches 12kW, liquid cooling is a more efficient choice to maintain the required operating temperature for each GPU and CPU. A single AI rack now produces over 100kW of heat, which needs to be effectively removed from the data center. Liquid cooling at data center scale can significantly reduce the electricity demand for a given cluster size. The power consumption can be reduced by up to 40%, allowing users to deploy more AI servers within a fixed power range, enhancing computing capabilities and reducing training time for large language models (LLMs), which is crucial for these large CSPs and AI factories."

What does the shipment of over 0.1 million GPUs mentioned above mean? Some comments suggest that if the average price of these GPUs is equivalent to the unit price of NVIDIA's $0.03 million chips, it could imply an order size in the billions of dollars.

Analyst Woo Jin Ho stated that assuming that 1,500 out of the 2,000 liquid-cooled racks claimed by Super Micro Computer were delivered in September, with each unit priced at approximately $3 million, the sales in September would be around $4.5 billion. This sets a good foundation for achieving the mid-point of the guidance range at $6.5 billion for Super Micro's fiscal first quarter of the 2025 financial year up to the end of September 2024.

Monday's announcement temporarily eased investors' concerns. Investors have been worried for months about Super Micro Computer's declining profit margin and suspicious accounting data. As of the closing price on Friday, October 4th, Super Micro Computer's stock price has fallen by more than 51% over the past three months.

Wall Street News mentioned that on August 28th of this year, Super Micro Computer announced a delay in submitting its 10-K filing to the U.S. Securities and Exchange Commission (SEC), the annual report for the company's fiscal year ending on June 30, 2024. They needed additional time to complete the evaluation of the design and operational effectiveness of the internal controls on the financial reports up to June 30.

Analysts believe that the timing of this delay is quite delicate, as just the day before, the well-known short-selling institution Hindenburg Research had announced shorting Super Micro Computer's stock, accusing it of having poor accounting practices, including undisclosed related-party transactions, sanctions, and export control failures.

Nearly a month later, on September 26th, after Hindenburg Research released the short report, the U.S. Department of Justice started investigating Super Micro Computer, with the investigation in its initial stages. A prosecutor from the U.S. Attorney's Office in San Francisco recently contacted individuals seeking information from a former employee who had accused Super Micro Computer and its CEO of accounting violations in April this year.

On September 26th, Super Micro Computer's stock fell by approximately 12.2% in a single day.

Analyst Ho mentioned that the delay in submitting the 10-K report and the investigation by the U.S. Department of Justice for Super Micro Computer may overshadow any 'favorable AI server demand fundamentals'.

'Lack of financial transparency may raise doubts about its sales momentum for the 2025 fiscal year. Following this trend (guidance), revenue needs to grow by 87%, which is the mid-point of its $26 billion to $30 billion guidance range. The company (Super Micro) has gained critical AI trades with its strengths in rack-level design engineering and shorter delivery times, but as financial distress signals emerge, concerns from customers and suppliers may escalate.'

Ho believes that the short report by Muddy Waters has highlighted many persistent risks for Super Micro that are already known to the public, summarized in Super Micro's 10-K filings, or have occurred before, such as delisting risks. It cannot be proven currently that Muddy Waters' investigation into Super Micro's financial operations is incorrect, but it is also difficult to directly link him to financial misconduct at Super Micro. However, Muddy Waters' report serves as a reminder that after experiencing exponential growth for several years, there is a need to increase transparency and strengthen financial controls.