Can Kioxia's listing in the US open up new capital entrances? A sharp rise of 456% during the year, pointing to the focus of global NAND configurations

The Zhitong Finance App learned that as analysts generally speculate that Kioxia, Japan's NAND memory chip leader, is about to knock on Wall Street and then go publicly traded in the US stock market like international semiconductor giants such as TSMC and SK Hynix, the memory chip market, which has recently returned to a bull market trajectory, has been catalyzed by both demand fundamentals and capital. The core investment significance of Kioxia's US listing plan as speculated by the market is to significantly expand the trading market where institutional investors can participate and increase funding sources for companies already driven by AI demand.

Kioxia's stock price has risen sharply by 456% in the Japanese stock market since this year, leading the Nikkei 225 Index, but Sebastian Thomas, investment manager of Voya Asset Management, an AI investment fund with a capital management scale of up to 14 billion US dollars, still does not hold the stock. The main concern is liquidity.

Thomas said that for large funds, in addition to corporate fundamentals, actual position opening capacity, transaction impact costs, exit convenience, and liquidity intensity also influence allocation decisions. Following the recent ADR listing of SK Hynix in the US stock market, it is expected to improve these conditions and increase Kioxia's investability in global institutional AI portfolios.

Memory chips broke through midsummer consolidation, and Goldman Sachs smelled a new round of storage bull market

Wall Street financial giant Goldman Sachs is once again optimistic about the bullish trading environment surrounding the memory chip sector. The core basis is that the expansion of demand for cutting-edge high-performance AI computing power brought about by the introduction of OpenAI Astra and the RSI (recursive self-improvement) training paradigm has begun to dominate the expected rise in demand for memory chips in the context of AI training. It has encountered low positions with traditional Wall Street asset management institutions and hedge funds. In particular, stocks such as Micron Breakthrough and Sandy began a downward trend in the summer technology line. The implicit volatility of semiconductors has declined, and the risk reduction in the risk gap previously opened by hedge funds has left room to re-increase their positions.

In terms of memory chip fundamentals, senior analyst Timothy Acuri from UBS predicts that the average sales price of memory chips in the third quarter increased by more than 20% in the context of a record high base in the second quarter, and that the shortage of DRAM and NAND will continue until 2027. The combination of bullish views on the two types of investment forms a bullish logic where “profit expectations are supported, and there is room to make up for low position capital”; whether it can be continuously revalued still depends on actual sales prices, shipments, and profit fulfillment.

Market performance before the market sell-off due to strong PPI in the US on Thursday already reflects this storage theme repair. Korea's KOSPI Index, which has the title of “AI computing power weather vane,” rebounded about 21% from its low level at the end of July in mid-August, reaching what is commonly referred to as the technical bull market threshold; it rose 4.6% on September 7, and Samsung Electronics and SK Hynix rose 5.7% and 8.1%, respectively. In the US market, as of the close of the US stock market on September 9, the Philadelphia Semiconductor Index had been rising for five consecutive trading days with a cumulative increase of nearly 6%. The Roundhill Storage ETF rose nearly 12% during the same period, indicating that the storage sector is leading the rebound.

An important industrial signal brought by Astra is that more complex jobs are beginning to have commercial value for execution by AI. According to media reports on September 10, OpenAI launched ChatGPT products for the financial services industry, combining GPT-6 Astra with professional data sources to support research, financial modeling, and customer material production. As deduced from this, the growth variables of AI demand will be further expanded to the number of concurrent agents, task execution time, tool call frequency, and context scale: when the cost of completing a task decreases and the success rate increases, enterprises have reason to deploy more workflows. This opens up huge room for cloud-based AI inference computing power and AI-related high-performance storage requirements. It is also the latest basis for the market to re-evaluate the sustainability of AI infrastructure growth.

From Tokyo stock market tycoon to Wall Street spotlight? Going public in the US may make Japan's Kioxia the focus of global AI investment

Looking at the underlying architecture, Kioxia's direct benefits are mainly NAND flash memory and enterprise-grade SSDs: agents call the enterprise knowledge base, retrieve vector data, load models, and save task status and audit records, all require persistent storage; long contexts and multi-round tasks will also expand the capacity requirements of reusable key-value caches, driving some caches to be stored in SSDs through a hierarchical architecture.

HBM undertakes high-bandwidth computing access, server DRAM undertakes low-latency work sets, and NAND provides more economical high-capacity storage. The three work together. Kioxia has positioned SSDs with high random read/write performance as an important product direction to improve AI inference efficiency. Relevant research has also shown that optimizing SSD cache scheduling can reduce repeated computation and GPU waiting. Therefore, Kioxia's long-term growth logic should focus on increasing enterprise-grade SSD capacity, performance value, and customer share; listing in the US improves capital market participation conditions, and continued profit growth still needs to be supported by product competitiveness, supply and demand, and orders.

The chief fund manager of a $14 billion fund focused on artificial intelligence investments said Kioxia's plan to go public in the US may make the Japanese chipmaker the focus of global investment again. Kioxia has soared 456% since this year, ranking first among the constituent stocks of the benchmark Nikkei 225 Index. People familiar with the matter revealed that the company previously stated at an internal meeting in May that it is preparing for a US stock market depository share (that is, US stock ADR) listing transaction to expand its investor base.

Voya Investments' Global Artificial Intelligence Fund does not hold Kioxia, nor does it hold any shares of technology companies that drive the Nikkei Index to rise. The fund contains $5 billion in funding from Japan.

Voya portfolio manager Sebastian Thomas said that the problem is liquidity, and trading in the US market may change this consideration, just like the boom in AI financing and investment that occurred after the Korean memory chip supergiant SK Hynix went public on the NASDAQ in July.

“There are many companies worth watching in Japan, especially those in the AI computing power infrastructure supply chain,” Thomas said in an interview with the media on Thursday. “The problem is finding companies that are sufficiently liquid to enable us to invest.”

According to information, this is an AI investment fund linked to Sumitomo Mitsui DS Asset Management. Since it was established ten years ago this month, the cumulative return has been about 600% based on the pre-tax distribution income reinvestment scale.

The fund focuses on investing in AI computing power infrastructure companies at the core of the computing power industry chain, such as NVDA.US (NVDA.US), AMD, and AI AISC leader Broadcom. Nvidia is its largest shareholder; it also invests in cutting-edge software application developers related to AI applications, as well as manufacturing companies that are expected to benefit from adopting cutting-edge AI technology, such as LY.US.

Thomas said that Voya has invested in Japanese companies in the past, and although it does not currently own Kioxia, it has investment exposure from other memory chip manufacturers, including SK Hynix and Micron Technology. He added that listing in the US makes this type of investment easier to carry out. “We generally prefer technology companies with better liquidity and larger positions and market capitalization.” he said.