Citrini's View: Terafab has completely excluded Chinese suppliers, and requires Korean firms to bid lower than Chinese vendors.

JPMorgan Chase estimated on Tuesday that most global banks have limited exposure to CoreWeave (CRWV.O), the Nvidia-backed cloud computing firm, after its stock and bonds plummeted last week amid investor concerns over its debt financing. While major U.S. investment banks and Japan’s Mitsubishi UFJ Financial Group may hold potential exposure to CoreWeave of between $1 billion and $3 billion, most European and Asian lenders have exposure of less than $500 million. Most of this exposure comes via delayed-draw term loans, which are primarily held through special purpose entities, with funds disbursed upon the achievement of project milestones. From the lending banks’ perspective, JPMorgan views the sector’s exposure to CoreWeave as well-managed and backed by robust collateral.

According to official announcements, Bitget will list Axis Robotics (AXIS) spot trading on its AI Zone. Deposits are now open, and trading will launch at 21:00 UTC+8 on October 21.

Beating AI News Flash: OpenAI’s 28-day product sprint is now on Day 15. Tibo first shared a ramen-cooking show titled "COOKED" with the caption "We're cooked" (literally meaning "we’re done for"). About 11 hours later, Tibo backtracked, posting: "Never mind, we’re still working on it… here’s another quota reset, enjoy." This round of reset has now taken effect.

Goldman Sachs released a new research report, maintaining its "Buy" rating for ChangXin Memory Technologies and slightly raising its 12-month target price from RMB 129 to RMB 130. The direct catalyst for the hike is ChangXin's disclosed third-quarter earnings preview: Goldman Sachs projects the company's Q3 revenue will range between RMB 115 billion and RMB 120 billion, marking a 15% to 20% quarter-over-quarter increase, with the midpoint roughly 30% higher than Goldman's prior forecast. This better-than-expected guidance, coupled with the company's announcement that its 5th-generation DRAM technology platform (G5) will enter mass production in September 2026, further reinforced Goldman Sachs analysts' confidence in ChangXin's medium- and long-term growth thesis. Accordingly, Goldman raised the company's profit forecasts for 2026 to 2030, and labeled it a core beneficiary amid China's DRAM demand upcycle. The company's current share price stands at around RMB 40, implying an upside potential of as high as 225% based on Goldman's target price. Notably, ChangXin's share has fallen sharply recently, with its current market capitalization dropping below RMB 3 trillion.

Crypto market analyst Axel Adler Jr. released his latest market observation, noting that Bitcoin has fallen around 6% from $86,700 on October 2 to $81,600 on October 20. The market is currently at a critical juncture for a structural correction, with bulls and bears battling near key support levels. According to Alex’s analysis, Bitcoin’s market structure signal has dropped from +0.37 a week ago to -0.32. Although the signal has moved out of the bullish range, it remains in the neutral zone and has not yet hit the bearish threshold below -0.5. He emphasized that the lower bound of the price range over the past three weeks is $80,400. As long as prices hold above this level, the current movement can still be considered a normal correction within the range; however, if prices break below $80,400 and the structure signal simultaneously drops below -0.5, the market correction will officially evolve into a downtrend. In the futures market, the index line reflecting the strength of long and short positions (buyers above 50, sellers below 50) has been below 50 since October 17. Although the index approached 50 on October 19, it failed to break through, and has now fallen back to 46. Alex stated that a decisive break above 50 for the index line will be the first signal of buyers returning. However, sellers currently hold absolute dominance, and futures market demand lacks sufficient momentum to push prices higher. Overall, Bitcoin is currently in a critical period of direction determination. The market needs to closely monitor the performance of the $80,400 support level and whether the futures long-short index can reclaim the 50 threshold, to determine whether this correction is a short-term adjustment or a precursor to a trend reversal.

Beating AI News (Dongcha) reports that last year, DeepSeek attempted to train its R2 model using Huawei Ascend chips but failed. The Financial Times previously disclosed issues with chip stability, inter-chip communication, and software support. Even with Huawei engineers on-site assisting, DeepSeek ultimately switched back to Nvidia chips to complete the training. Over a year later, the two parties have deepened their collaboration to include hardware design. The Information now reveals that Huawei engineers are permanently based at DeepSeek’s offices, integrating the practical demands of model training directly into the design of next-generation Ascend chips—including how to schedule computing tasks and transfer data between chips. At the end of September this year, DeepSeek open-sourced the Ascend version of TileLang, plus underlying tools for matrix computation, MoE expert communication, etc., to reduce the development work required for migrating models to domestic chips. However, getting the software to run is still a long way from full pre-training. Tens of thousands of chips need to operate continuously for weeks; any single chip failure or data transmission delay could slow down the entire cluster. Now, the two sides are starting to tackle these issues at the chip design stage. Huawei plans to prioritize allocating most of its supply of Ascend 960DT, set to be ready in Q1 2027, to DeepSeek rather than distributing it to other model companies. DeepSeek, for its part, is preparing to start pre-training as early as the end of this year using the 950DT. So far, related tests have used over 1,000 chips, but full pre-training may require tens of thousands, and some cluster management software remains incomplete. Meanwhile, DeepSeek is still seeking more Nvidia chips.