analysis

CryptoQuant analyst Zizcrypto stated that the Stablecoin Supply Ratio Oscillator (SSR Oscillator) 90-day reading quickly rose to the "strong stablecoin buying demand" zone, reaching 3.74 on August 21, close to the peak of about 4 expected in November 2024. Since then, the indicator has started to decline, while BTC is currently trading around $77,000.The rapid rebound of the stablecoin demand indicator coincided with a strong liquidity pulse that occurred earlier in the market. However, what is more concerning now is not the short-term spike itself, but whether this demand signal can be sustained. If the 90-day SSR Oscillator continues to decline and falls back below the "high" zone, the confirmation strength of this indicator regarding the sustained increase in stablecoin buying demand will weaken. Currently, this change appears more like a liquidity pulse rather than a formed persistent demand expansion cycle.The demand signals from the stablecoin side have indeed strengthened significantly recently, but the 90-day oscillation indicator has already started to cool down from its peak of 3.74. If it cannot maintain itself in the "high" zone, then whether this liquidity improvement can evolve into more lasting demand growth remains in question.

User Wordgrammer stated that the internet has collapsed due to social and video being 100% sandboxable, with consumer preferences leaning towards usage while intoxicated, in bed, on the toilet, or when 99% of their brain is occupied. They believe that the current AI discussion is similar to the misjudgment in 1995 that "everyone will have a website," as not everyone will create personal software or mobile applications.Wordgrammer indicated that everyone will use agents, but less than 1% will use coding agents, just as everyone has a social homepage but less than 1% have a personal website. Consumer-grade and enterprise agents must have a 0% failure rate foolproof sandbox, where accidental purchases, messaging, or information leaks cannot occur. Better models are insufficient to achieve this and require a rethinking of abstraction.

According to CoinDesk analysis, Bitcoin is approaching the "golden cross" technical signal where the 50-day moving average crosses above the 200-day moving average. Historical data shows that since 2012, Bitcoin has experienced 12 golden crosses, with an average increase of 24.9% over three months for 9 of them, but only 3 have maintained for a full year without being broken by a death cross, with an average annual increase of 250%.What makes this golden cross different from previous ones is that the USDT market share is simultaneously approaching a "death cross" ------ its 50-day moving average is about to cross below the 200-day moving average. A decline in USDT market share usually indicates that funds are flowing out of stablecoins and into risk assets like Bitcoin, and this indicator has previously marked significant trend changes for Bitcoin. The combination of both further strengthens the bullish signal for this round of Bitcoin rebound.

Rapha Zagury, CEO of Twenty One Capital and founder of Elektron Energy, stated during his speech at Bitcoin Asia 2026 that the Bitcoin network is experiencing its first-ever bear market in hashrate. The hashrate of the Bitcoin network was close to 1.3 ZH/s at the end of last year, but has since been slowly declining, with the duration of this decline from the historical peak now setting a record. Zagury believes that Bitcoin mining is not simply a "good business" or "bad business"; it largely depends on where the mining company stands on the cost curve. Mining companies with lower energy costs and higher machine efficiency can maintain higher profit margins, while those with high energy costs and low equipment efficiency may be forced to shut down.Currently, while the Bitcoin hashrate price has improved compared to before, it is still at a relatively low level when measured against historical standards. When the price of Bitcoin rises faster than the growth of the network's hashrate, mining is more likely to outperform BTC. For companies, he believes that the best risk-adjusted allocation is not simply choosing to "buy BTC" or "mine," but rather a combination of both; however, if only $1 can be allocated, he suggests prioritizing the purchase of BTC. Regarding energy issues, Zagury stated that energy consumption itself does not imply waste; energy is the foundation of economic development and human progress. He believes that one of the greatest characteristics of Bitcoin mining is its highly flexible load, as mining machines can quickly turn on and off based on energy supply, thus helping the grid absorb idle or surplus electricity and enhancing grid stability to some extent. Additionally, he believes that Bitcoin mining is generating "option value" that was not previously apparent, including aspects such as energy utilization, market share, proximity to the Bitcoin protocol, and infrastructure. With the growing demand for AI and high-performance computing (HPC), the existing energy and data center infrastructure of mining companies may also gain additional application scenarios such as AI computing power. Currently, among large publicly listed mining companies, there are fewer and fewer that can continue large-scale Bitcoin mining, and the industry is at a critical stage where the energy revolution intersects with the Bitcoin revolution.

According to Cointelegraph, the global bond bear market continues to ferment, with Japan's 10-year government bond yield rising to 3% on Tuesday, the first time since 1996; the 30-year government bond yield also broke through the historical high of 4.18%. The U.S. 10-year government bond yield simultaneously rose to a multi-year high of 4.78%, and global long-term sovereign bond yields are at their highest level since the 2008 financial crisis.In this context, Bitcoin remains in a sideways consolidation, maintaining around $78,000, slightly retreating from an earlier high of nearly $79,000. There is a dense resistance area between the current spot price and $86,000, which limits Bitcoin's upward momentum. Market sentiment remains cautiously optimistic in the short term, with the $76,000 to $82,000 range seen as a key battleground in the coming weeks.This round of selling occurred after U.S. Treasury Secretary Yellen announced an increase in the upper limit of government bond repurchase transactions to $4 billion starting in September, with some commentators likening it to a form of yield curve control. Arthur Hayes has long argued that the Federal Reserve will eventually activate the FIMA repo facility, a mechanism that will create new dollar liquidity, which is also why he recommends allocating Bitcoin, gold, and cryptocurrencies; Yellen hinted at the future use of this tool as early as August.

Greeks.live researcher Adam stated: "I see some people discussing the Strategy to restart buying coins. This short-term stimulus does not affect the long-term trend, and the Strategy is also a big trap. BTC now has no possibility of market manipulation, and no one can predict the trend. It can only be said that after the Strategy has floating profits, there is more room to buy coins. Looking at the long term, the overall ETF is still flowing out, and it is very difficult to stabilize at $80,000, with significant selling pressure.On the macro front, whether it is the Federal Reserve's hawkish remarks or the escalation of the US-Iran conflict, both are challenging the already fragile investors. The short-term volatility is decreasing rapidly, and it is difficult to support a bull market with just one bullish candle. I hope that after a slight correction, it can reach above $85,000; otherwise, arbitrageurs and sellers will create significant selling pressure. The Gex at the end of the month is gathering, and we need to maintain this momentum."

CryptoQuant analyst Axel Adler Jr stated that from August 18 to 28, the 30-day cumulative distribution of Bitcoin long-term holders (LTH) increased from 174,500 to 281,900, a rise of 61.5%, reaching the highest level since 2026 on August 28. This metric accelerated significantly during the rapid rebound after the short-term squeeze in Bitcoin, indicating that long-term holders are taking profits as prices rise.Meanwhile, the LTH MVRV rose from 1.31 on August 18 to 1.64 on August 27, remaining at a level of 1.6 on August 31, which means that the market value of Bitcoin held by long-term holders is approximately 60% higher than their average realized value. This suggests that despite the increase in distribution activity, long-term holders still possess significant unrealized profits. Axel believes that the price rebound after the short-term squeeze has simultaneously elevated the profit levels of long-term holders and the speed of token distribution, creating conditions for further profit-taking. If Bitcoin prices remain stable, long-term holders may continue to release supply, adding extra selling pressure to the market. The current key issue for the market is whether the high-level distribution by long-term holders will continue and whether new buying can absorb this additional supply; otherwise, the market may face price pressure again.

Venture capitalist Tomasz Tunguz posted on X that the cutting-edge AI market is splitting into closed camps, with labs choosing partners, cutting off competitors, and rationing access to the strongest models. Salesforce selected Anthropic as its dedicated AI partner, making Claude the default model for the world's largest CRM and Slack.OpenAI cut off its API access on November 12 after SpaceX acquired Cursor. Z.ai placed its flagship model under a security review after cumulative revenue exceeded $10 billion. Anthropic rationed Mythos 5 through Project Glasswing and restricted Fable to inference only in the United States. OpenAI first launched the GPT-5.6 variant to trusted partners.Nvidia is investing in an open ecosystem, including $13 billion for Hugging Face, $7 billion for Poolside, and a $26 billion commitment to Nemotron. Tunguz stated that access, rather than price, has become the new scarcity at the forefront.

SemiAnalysis released the latest episode of its podcast, which pointed out that SK Hynix and Micron have been facing challenges in achieving the highest HBM4 speeds. In particular, SK Hynix had to redesign its base chip, resulting in delays in HBM4 deliveries to the NVIDIA Rubin platform.Samsung currently has the best HBM4 technology. Broadcom has traditionally relied heavily on Samsung for its HBM supply. Samsung's performance in the HBM3E generation was poor, which posed a disadvantage for Broadcom's ASICs. However, now Samsung is leading its competitors with HBM4, and this same supply chain relationship has become an advantage for Jalapeño.

Blockchain analysis company Chainalysis has filed a lawsuit in the U.S. Federal Claims Court, accusing the U.S. Immigration and Customs Enforcement (ICE) of illegally awarding a $94.6 million exclusive contract to competitor TRM Labs. This contract provides forensic software and support services for investigations by the Department of Homeland Security's task force, with a performance period from July 1, 2026, to June 30, 2027. Chainalysis claims this is the largest blockchain analysis contract awarded by the U.S. government to date.In the complaint, Chainalysis pointed out that ICE issued a Request for Information (RFI) on May 28, which included 18 questions regarding capabilities such as a proprietary fraud victim database, AI agent data retrieval, automatic notifications for virtual asset service providers (VASP) to execute voluntary freezes, and collaboration with stablecoin issuers. However, many of the RFI requests did not appear in the final Statement of Need released less than two weeks later, while ICE's market research determined that Chainalysis could not meet the needs based on RFI standards. Chainalysis also noted that "sexual extortion," which was not mentioned in the RFI, became one of the three main task areas in the Statement of Need, and some requirements closely matched TRM's proprietary product architecture.Chainalysis claims it was the only company to submit a capability statement, but ICE completed its market research the next day, determining that TRM was the only vendor capable of meeting all the requirements, and did not pose any follow-up questions to it.

The semiconductor and AI independent research organization SemiAnalysis released a deep security report on Neocloud (new cloud), revealing various cross-tenant security vulnerabilities discovered during the ClusterMAX 3 testing period. In a four-month test covering 25 vendors and 32 clusters, the team achieved multiple instances of cross-tenant remote code execution (RCE) solely by exploiting publicly known vulnerabilities and basic configuration checks. Affected entities included banks, telecommunications companies, universities, research institutions, AI laboratories, and even a national intelligence agency.Typical issues included: shared Kubernetes control plane leading to tenant metadata visibility, container escape, exposure of BMC/IPMI management networks, incorrect configuration of InfiniBand security keys (P_Key, SA_Key, M_Key), unfortified default trust mode of BlueField DPU, Grafana monitoring dashboards using god-level API keys, and lack of VXLAN isolation in front-end networks. The report specifically pointed out a cascading vulnerability case: a misconfiguration of shared vCluster combined with software versions being two years out of date ultimately completed the POC verification of cross-tenant RCE within an afternoon.Notably, the report questioned the mainstream narrative that "AI has fundamentally changed the pace of cybersecurity": statistics on CVEs for NVIDIA GPU drivers, CUDA, PyTorch, Kubernetes, Docker, and the Linux kernel showed that there was no significant increase in vulnerabilities after the popularization of AI coding models, with most data supporting the "no change hypothesis." The report also detailed the incident where an OpenAI-trained agent attacked Hugging Face, where the AI agent achieved cluster-level privilege escalation through a message board established via Artifactory, which went undetected from May to July. While building POC verification for existing vulnerabilities, the team found that Claude Fable and GPT-5.6 Sol frequently rejected security-related requests, ultimately relying on open-source models such as DeepSeek V4, Kimi K3, and GLM-5.2 to complete the task.SemiAnalysis stated that the core issue in the Neocloud (new cloud) industry is not the new risks brought by AI, but rather the long-term absence of basic patch management, tenant isolation, and security design. They recommended that vendors establish automated security announcement monitoring systems and rectify single points of failure that could expose all users' architectural patterns.

According to CoinDesk, as Bitcoin surged from about $62,000 to over $77,000 last week, liquidating approximately $3 billion in leveraged short positions, major crypto market makers such as Abraxas Capital, Fasanara Capital, and Wintermute quietly established short positions in perpetual contracts worth hundreds of millions of dollars on Hyperliquid.Lookonchain data shows that the three institutions collectively hold short positions of 138,569 ETH (approximately $338 million) and 3,425 BTC (approximately $265 million); meanwhile, Abraxas Capital withdrew 73,872 ETH (approximately $173 million) from Binance in the past four days.This strategy is known as cash arbitrage or basis trading: traders hold spot positions while shorting an equivalent amount of assets through perpetual contracts to hedge against price volatility risks, primarily earning the funding rates paid by longs to shorts. In previous months, funding rates were long suppressed or even turned negative, but this month's rebound has quickly turned funding rates positive, reopening the arbitrage window. Aegis data shows that the 30-day average funding rate for Bitcoin perpetual contracts on August 24 reached an annualized 6.7%, with a 7-day average of 8.7%; 21shares capital markets noted that basis trading for mainstream assets like Solana is also becoming lucrative.This trading has extended to regulated markets: Glassnode data shows that CME Bitcoin futures open interest has risen from about 87,000 BTC to 122,000 BTC.

According to Cointelegraph, Bitcoin failed to effectively break through the $80,000 mark, with BTC/USD dropping to $78,442 on the Bitstamp platform, a decline of about 1% for the day, after Federal Reserve Chairman Kevin Walsh delivered a cautiously toned speech in Jackson Hole. In his speech, Walsh reaffirmed the Fed's commitment to the 2% inflation target and downplayed the significance of the recent decline in CPI and PCE data.Walsh stated that although these broad inflation indicators have significantly retreated from their highs a few years ago, the progress over the past two years has been relatively modest. The better-than-expected PCE and CPI readings this summer do not indicate that the underlying trend has materially improved. He also noted that the forward guidance, a conventional practice introduced during the financial crisis, "is now outdated" and will not return in the future. U.S. stocks were not weighed down by his remarks, with the S&P 500 and Nasdaq indices both rising by about 0.5%.On-chain data shows that there is a dense resistance zone between the current spot price and $86,000, which suppresses upward momentum. QCP Capital analysis pointed out that even if the price breaks through, the derivatives market needs to support it by controlling funding rates and the growth rate of open interest. The key is whether the subsequent market movement is driven by spot participation or leveraged positions. As of the time of writing, BTC/USD has risen 26.35% this month, marking the best August performance since 2017.

Analysis of the AI semiconductor and infrastructure cycle indicates that the demand for AI infrastructure will continue to exceed that of the internet era, as user penetration and per capita token are multiplied and converted into tokens, significantly raising the ceiling. User penetration has surpassed 50%, with growth primarily coming from the still-early per capita token; the median monthly AI spending per employee in U.S. companies is about $12, which could long-term approach around 10% of white-collar salaries, approximately $1,000 per month, leaving nearly two orders of magnitude of space in between. Unlike the flat subscriptions and extremely low marginal hardware consumption of the internet, the high costs of inference make the marginal cost of a single access higher, requiring greater infrastructure intensity.After developers program, the next ARR growth will still mainly come from broad programming: non-programmers use programming infrastructure to complete non-programming tasks across industries, with programming becoming the default execution kernel for agents. Tasks related to broad programming account for about 60% to 70% of ARR. As of June 2026, Codex accounted for 64% of the total output tokens from Codex and ChatGPT among OpenAI's enterprise clients; since February, Codex has seen a much higher weekly growth in verticals such as law, sales recruitment, and marketing compared to engineering. In Anthropic's revenue, narrow development/software accounts for about 40%, while finance and insurance exceed 20%, with law, life sciences, retail, and others also having considerable shares.The demand-side token growth logic remains, with a high overlap between funders and beneficiaries.

CryptoQuant published an analysis stating that the Bitcoin Coinbase premium index rebounded to 0 on August 26 after being negative for 14 consecutive days. However, this rebound was accompanied by large-scale profit-taking: last week, realized profits averaged $933 million, a week-on-week increase of 824%; exchange inflows for 1 to 3-month tokens increased by 280%, and daily inflows to wallets holding 1,000 to 10,000 BTC reached 1,136 BTC (+235%).There are significant differences in exchange traffic: Binance had a net inflow averaging +2,184 BTC, while the overall exchange net outflow was -2,372 BTC, with Coinbase experiencing a single-day outflow of 3,499 BTC coinciding with the day the premium returned to zero. Long liquidations increased by 469%, far exceeding the 15% for short liquidations, and stablecoins continued to flow out, indicating that buying support is weakening. Analysts point out that history shows that under this structure, the probability of range-bound fluctuations or shallow corrections is higher than direct increases, and sustained upward movement requires the premium to remain above zero and stablecoin net inflows to turn positive; currently, only the first condition is met.

P Equity Research and SemiAnalysis researcher Nick Doyle and others discussed high bandwidth flash (HBF) in X Space. Nick stated that it is still too early to determine how much the cost premium of HBF relative to HBM can shrink; existing data mostly comes from vendor claims, such as Sandisk stating that the cost per bit is about one-eighth that of HBM. Yields, testing, and other factors will improve with scale, but structural costs such as TSV, stacking, and pSLC mode will always exist, and durability is a key unknown; if wear exceeds expectations, costs will rise.The application scenarios for HBF are narrow, targeting only AI inference, especially low batch and long context MoE models, and it is not a substitute for HBM. The actual bandwidth target is about 1.6 TB/s, which is at the HBM3E level, suitable for sequential reads to load model weights, more aligned with the capacity needs of a small number of GPUs in local or private enterprises, rather than ultra-large-scale bandwidth scenarios. Heat dissipation reliability has not yet been resolved; flash memory will degrade faster at high temperatures next to GPUs, and mitigation measures such as UCIe separation and daily refresh have yet to be validated.In terms of manufacturing, Sandisk/Kioxia has experience with 3D NAND, and SK Hynix complements HBM-style stacking capabilities, but mass production is still to be confirmed. Overall, storage is shifting towards a specialized layered market, with NAND shortages expected to continue until 2028, and HBF may further impact supply and demand.

According to CoinDesk, Bitcoin is currently stable around $80,000, with traders focusing on Federal Reserve Chairman Kevin Walsh's keynote speech at the Jackson Hole annual meeting. The annual seminar hosted by the Kansas City Fed opened on Wednesday, and Walsh will speak on Friday morning. Current inflation remains well above the Fed's 2% target, and long-term borrowing costs are near multi-year highs, with the Fed maintaining the benchmark interest rate range at 3.5%-3.75% in recent months.Mark Connors, Chief Investment Officer of Risk Dimensions, expects Walsh to keep the option of raising interest rates but not to raise rates before the midterm elections. The July PCE inflation rate reached 3.7%, partly driven by rising energy prices following the Iran war. Samir Kerbage, Chief Investment Officer of Hashdex, stated that Bitcoin does not directly respond to the September interest rate decision but follows global liquidity and long-end yield curve trends, similar to the drivers of gold. Last week, the U.S. Treasury announced an increase in long-term Treasury bond purchases after the 30-year Treasury yield hit a 19-year high, which pushed Bitcoin higher; Kerbage described this surge as "primarily a liquidity event."This year's Jackson Hole meeting focuses on financial innovation, payments, and policy, with stablecoins, tokenized deposits, and faster settlement systems becoming the discussion highlights. Kerbage believes that if Walsh views these technologies as part of the financial system rather than risks to be curtailed, the impact could extend beyond Bitcoin, with smart contract networks and protocols handling tokenized payments and settlements benefiting more directly from this shift.

The "Realized Cap Impulse" indicator for Bitcoin has recently ended a continuous 93-day negative state and has turned positive for the first time, marking the longest reversal signal after a capital contraction cycle since the bear market of 2022.This indicator measures the momentum of changes in the realized market capitalization by tracking the changes in realized cap, combined with factors of Bitcoin supply and price, to assess whether the flow of tokens with actual economic significance in the market is driving capital base expansion. The positive shift in the indicator does not merely reflect a price increase but indicates that the flow of funds within the Bitcoin network is changing.Data shows that the indicator broke above the zero axis on August 20 when the BTC price was around $73,000 and has maintained positive values for 8 consecutive days. Currently, the BTC price has risen to about $78,900, an increase of approximately 8% during this period.As of the latest, the indicator reading is 0.198, below the peak of 0.226 reached on August 26. Historical data indicates that similar signals at the end of bear markets have been accompanied by significant rebounds in Bitcoin: after the indicator turned positive in September 2015, Bitcoin rose about 160% within a year; after March 2019, it increased about 173% in 90 days; and after January 2023, it rose about 45% in 90 days, with a yearly increase of 104%.However, the realized cap momentum indicator is not an absolute signal of a cycle bottom. Similar positive shifts occurred in early 2018 and 2022, but the market did not immediately enter a sustained upward phase afterward. Analysts believe that the indicator's continued positive value, confirmed by price trends, is more valuable than a single-day breakthrough above the zero axis.

In the latest podcast, SemiAnalysis founder Dylan Patel predicts that by 2028, OpenAI and Anthropic may account for 70% to 80% of the world's new AI computing power, with the total computing power scale potentially exceeding 100GW. Patel stated that the two companies currently account for about 30% of the world's annual new computing power, and this proportion is still rising rapidly.Patel pointed out that the business model of leading AI laboratories is changing, with a significant increase in the efficiency of AI computing power output. Currently, Anthropic's revenue per megawatt of computing power has reached about $50 million and may further rise to $100 million. This allows OpenAI and Anthropic to procure or lease computing power at high prices ranging from $25 million to $50 million per megawatt.Patel expects that global AI-related capital expenditures will reach about $11 trillion from 2024 to 2029, with over $5 trillion needing to be financed through debt. Due to the potential return on investment of AI infrastructure being far higher than that of traditional industries, tech giants may accept higher financing costs, thereby pushing up overall credit rates and squeezing the valuations of traditional assets and highly leveraged economies. Additionally, Patel believes that the new computing power may not primarily be used for providing model inference services externally, but may instead flow more towards internal research and development and self-improvement of models within AI laboratories.