NVIDIA discloses holding approximately $21 billion worth of shares in SpaceX, further deepening capital ties within the AI industry.

Semiconductor stocks regained buying interest after a sharp sell-off in July, and Bank of America’s (BofA) latest analysis further reinforces this rebound thesis. The U.S. lender notes that demand for AI computing and memory is stronger than previously anticipated, and there remains upside potential for the growth trajectory of the AI server chip market, with AMD and NVIDIA standing as the most representative top picks. BofA’s core view is that the market had earlier over-worried about cloud providers slashing AI capital expenditures, but recent earnings reports and guidance indicate that AI infrastructure investment is still ongoing. Cloud giants including Microsoft and Amazon continue to signal robust demand, with recovery observed across the server, GPU, networking, storage, and power supply chains. The bank favors AMD because its AI accelerator and server CPU portfolio will benefit from customer diversification; it remains bullish on NVIDIA, as the firm’s product ecosystem, supply chain control, and next-generation platforms will sustain their industry-leading positions.

According to an official announcement from Bitget, the platform’s stock dual-currency investment product has officially launched, with its first batch covering over 20 popular U.S. stocks and ETF-related underlying assets including rNVDA, rTSLA, rAAPL, and rMETA. This upgrade adjusts the product’s settlement time to 23:30 (UTC+8), aligning more closely with the post-opening trading sessions of U.S. stocks. In addition, Bitget has rolled out two limited-time promotions simultaneously. During the campaign period, new users participating in Bitget’s dual-currency investment for the first time who complete designated tasks such as net deposits will be eligible to receive a dual-currency experience bonus, with a maximum reward of 3,000 USDT. This promotion ends on August 21. Meanwhile, users who purchase dual-currency products during the campaign will also get limited-edition Bitget merchandise including camping sets and commemorative gold coins, with this offer running until August 28. For more details, please visit Bitget’s official website.

According to HTX market data, PUMP has gained 19.65% over the past seven days and 66.57% over the past 30 days, with its market capitalization now standing at $1.665 billion. This morning, the monthly unlock of tokens held by the Pump.fun team and its investors was completed, releasing a total of 4.94 billion PUMP tokens valued at approximately $13.6 million. The unlocked tokens were subsequently distributed to 125 wallet addresses.

Pumpfu(@Pumpfun) unlocked another 4.85B $PUMP($13.6M) to 124 today.

According to monitoring by crypto analytics account Ai Yi, an investor encountered an extreme "paper wealth" scenario after participating in the LAB token public offering. Crypto community user Skylinee joined LAB’s public offering in October 2025, investing $5,000. The token’s value once surged to approximately $5.6 million, marking a 1,120x increase within nine months. However, due to the project team unilaterally delaying the token unlock schedule, the investor only received the tokens recently. Data shows that the current value of his held LAB tokens is just around $3,219, a roughly 99.94% plunge from its previous peak.

Western Digital’s latest analysis notes that as AI applications scale rapidly, AI data center construction is shifting from a pure GPU computing power race to a competition in data storage capacity, with storage planning becoming a core component of AI infrastructure. Citing IDC forecasts, global annual new data volume will reach 718 ZB by 2030. Data generated by AI systems does not disappear once computing tasks end; training data, model checkpoints, embedding vectors, inference logs, prompts, output results, and evaluation data will all accumulate continuously. Western Digital points out that many current AI infrastructure plans overemphasize GPU utilization while neglecting data accumulation throughout the AI lifecycle. Data generated during training and inference will become critical assets for model iteration, quality assessment, and compliance audits, with storage costs directly impacting the long-term operational efficiency of AI systems. As data scales to PB and even EB levels, a single storage architecture can no longer meet demands. Enterprises need to adopt a tiered storage strategy: using high-performance flash memory for training and real-time inference, and high-capacity HDDs and object storage for long-term data archiving, historical records, and low-frequency access scenarios. The analysis concludes that the key metrics for future AI infrastructure competition will not only be the number of GPUs, but also per-PB data storage cost, energy consumption, recovery efficiency, and data lifecycle management capabilities. If enterprises still treat storage as an afterthought to computing, they may face issues such as out-of-control data costs and reduced model iteration efficiency.