Did Anthropic's Below-Expected Annual Revenue Spark a Market Sell-Off? U.S. Pre-Market Major Indices, Storage and Optical Communications Sectors See Broad Losses.

On-chain analytics firm Glassnode says the Bitcoin market is in a "highly contracted" state, with selling pressure from sellers easing but buyers yet to enter noticeably. BTC is currently fluctuating mainly between $63,000 and $68,700, with market trading activity dropping to its lowest level since 2019. Glassnode data shows the 30-day Seller Exhaustion Constant has fallen to its lowest level since 2013, though a historic market bottom signal has not yet been confirmed. If BTC breaks below $58,500, the next support level could drop to $52,800. Crypto analyst CryptoBusy notes that the current degree of seller exhaustion is approaching levels seen in previous bear market bottoms, though deeper sell-offs have occurred in historical cycles, meaning Bitcoin may still need to release more selling pressure to complete market clearing. Michael Terpin, founder of Transform Group, predicts Bitcoin could bottom around $57,000 in October, and views complete seller exhaustion as a key signal for the end of the bear market. However, there is no consensus on market bottoming among analysts. Some analysts point out that only 2 of Glassnode’s 10 capitulation indicators have been triggered so far, putting BTC’s potential bottom at between $49,000 and $53,000. Analysts also add that before ETF inflows and spot trading volumes show a clear recovery, it is not enough to confirm the market has entered a sustained accumulation phase. Glassnode further warns that leveraged traders have already bet on a market recovery, but current on-chain and market data have not sufficiently supported this expectation.

Bank of America’s latest Global Fund Manager Survey shows that global fund managers’ allocation to stocks has risen to its highest level in nearly five years, with market consensus becoming extremely crowded. A net 56% of respondents are overweight stocks, the highest since November 2021, while cash holdings dropped to a historic low of 3.5%. Bank of America Chief Investment Strategist Michael Hartnett noted that the current market has formed a highly uniform expectation of “no macroeconomic landing, no Federal Reserve rate hikes, no AI capital expenditure cuts, no Democratic landslide, and no short sellers.” He believes that current positions are more suited for retreat or rotation within risk assets rather than expanding overall risk exposure. The survey shows 72% of respondents expect the Fed to hold off on rate hikes ahead of the November midterm elections, and 71% anticipate no cuts to AI capital expenditures by hyperscale cloud computing firms this year. However, the AI bubble has been identified as the biggest tail risk, with hyperscale cloud computing firms’ capital expenditures seen as the most likely trigger for a credit event. Meanwhile, Jonathan Krinsky, Chief Market Technical Strategist at BTIG, warned that August 18 to October 11 is typically one of the toughest periods for markets in U.S. midterm election years. Historical data shows that since 1990, with the exception of 2006, the S&P 500 has fallen by at least 7% between August and October in every midterm election year. The S&P 500 has risen more than 13% year-to-date and is trading at a historic high, while U.S. 10-year and 30-year Treasury yields have climbed above 4.7% and 5.2% respectively. Rising energy prices and financing costs could also further pressure the stock market. Krinsky advised investors to reduce risk exposure or hedge during this historically high-risk window.

As crude oil prices pull back, refined product prices continue to climb. The U.S. diesel-crude oil crack spread has surged to $102.20 per barrel, hitting an all-time high. Disruptions to global energy supplies from the Iran and Ukraine conflicts have tightened diesel supplies, while the ongoing agricultural harvest season has boosted demand for fuel for farm equipment like tractors, further driving up diesel prices and potentially passing those costs to inflation via food, transportation, and heating expenses. Meanwhile, WTI crude oil prices have broken through the downtrend line formed since their April peak, signaling the end of the four-month downward trend. Should oil prices rise further, market inflation expectations could reheat. Energy prices and inflation risks, paired with government debt concerns, are pushing up yields on U.S. Treasuries and other developed-market bonds, raising the opportunity cost of holding risk assets like Bitcoin and potentially limiting BTC’s upside potential. However, a weaker U.S. dollar is still providing some support for Bitcoin. The U.S. Dollar Index fell to 99.29 on Monday, hitting a two-and-a-half-month low and breaking below its prior uptrend line. Currently, Bitcoin is being influenced by multiple factors including oil prices, bond yields, and the U.S. dollar’s movement, with the macro environment showing clear divergence.

Blockchain analytics firm Chainalysis' government division, Chainalysis Government Solutions, has filed a lawsuit against U.S. Immigration and Customs Enforcement (ICE) over its direct award of an approximately $94.6 million contract to rival TRM Labs, challenging the procurement decision as "unreasonable and arbitrary." Chainalysis filed a protest with the U.S. Court of Federal Claims on July 27 regarding the one-year contract, which runs from July 1, 2026, to June 30, 2027, to provide blockchain forensics software and support services to a homeland security task force. TRM Labs intervened in the case on July 28 to defend the contract jointly with the U.S. government. ICE conducted market research in just about six days, received responses from eight companies, and concluded only TRM Labs had the required capabilities. Chainalysis’ core claims remain sealed due to trade secret concerns. The court has scheduled an oral argument for September 2 and ordered the government to issue a decision by September 10 at the latest.

Affected by the continuous surge in storage chip prices, Apple plans to complete price adjustments for the iPhone 17 series across all global regions by the end of August, with the maximum per-unit increase likely not exceeding 1,000 yuan (RMB). The specific plan may reference the previous price hikes in the Japanese market. Apple raised prices for its entire iPhone lineup in Japan on July 17, with increases ranging from 8% to 11.3%. Among them, the iPhone Air saw the highest hike at 11.3%, the iPhone 17 Pro Max rose by 10.3%, adding 20,000 yen per unit. Calculated at the exchange rate at that time, the cost of purchasing a single iPhone in Japan increased by approximately 540 yuan to 835 yuan. Apple has not yet announced price hikes for the iPhone 17 in the Chinese market, but it raised prices for Mac, iPad, and other products in June, with some models seeing increases of 15% to 20%. Apple CEO Tim Cook previously stated that storage chip prices continue to rise, and the company is evaluating response measures. He described the current memory price environment as a "once-in-a-hundred-year flood," noting that prices have risen exponentially.

Anthropic’s annualized revenue run rate as of the end of July has surpassed $65 billion, a nearly 40% increase from $47 billion in May, and more than seven times the roughly $9 billion it reported at the end of 2025. The company’s preliminary Q2 revenue exceeded $11.5 billion, with a year-over-year increase of at least 14x. Reuters previously reported that Anthropic has set a revenue target of $190–$200 billion for 2028, with market discussions around its potential IPO valuation exceeding $2 trillion, and some investors even deeming a $3 trillion valuation within reach. If Anthropic hits $200 billion in revenue by 2028 and trades at a 50x price-to-sales ratio, its theoretical market capitalization would reach $10 trillion. This outlook is underpinned by a growing "model-computing power-capital" cycle taking shape in the AI industry. Giants like NVIDIA and Amazon are continuously providing computing power to leading AI labs via investments, financing, and data center construction, while model companies such as Anthropic and OpenAI have become core demand drivers in the entire AI infrastructure investment chain. Notably, Anthropic’s growth rate is catching up to and even surpassing OpenAI’s. At the end of 2025, OpenAI’s annualized revenue stood at roughly $20 billion, more than double Anthropic’s; as of the end of July, Anthropic’s annualized revenue hit $65 billion, while OpenAI’s latest annualized revenue is around $40 billion. While a $10 trillion market cap remains an extreme hypothetical and faces risks including competition, computing power costs, and valuation corrections, as AI expands from infrastructure to the "intelligent productivity" layer, Anthropic has emerged as a key candidate for the market’s bet on the next company to top $1 trillion—and even $10 trillion—in valuation.