On July 29, 2026, the information flow seemed to be split in half: on one side, the cryptocurrency market continued its slump since 2025, with the media focusing on the disclosed second quarter spot trading volume of the top ten centralized exchanges — dropping from $2.70 trillion in the first quarter to $1.95 trillion, a sharp decrease of 27.9%. Even though Binance still holds about 38.7% of the market share and Bybit about 10%, it couldn't stop the narrative of "declining volume and cold prices" from continuing to brew; on the other side, the AI sector was simultaneously putting on drastically different performances: Meta CEO Mark Zuckerberg publicly opposed the idea of "banning the use of advanced Chinese AI models in the U.S.," highlighting internal government divisions on how to manage Chinese models. OpenAI launched GPT Transcribe and GPT Live Transcribe in the direction of voice transcription and reduced related model prices by about 25% to capture the voice entry at lower costs, while Ant Group's database company OceanBase was rumored to be gearing up for its first round of external financing and responded, "We are fully committed to innovation in data technology and products for the AI era," forming a complex backdrop together with semiconductor stocks — SK Hynix, Micron, SanDisk, and Nvidia — which continued to decline in pre-market trading. One side saw genuine on-chain trading volume retreating, while the other side was intensifying its capital stories around models, chips, and data infrastructure, presenting a fragmented picture where "money is tilting towards AI, but the direction and rhythm are far from solidified."
$1.95 trillion Shrinks: CEX Heat Hits the Brakes
Bringing the lens back from AI to the market, the central figure for Q2 2026 was glaringly stark: the spot trading volume of the top ten centralized exchanges had shrunk to about $1.95 trillion, down from $2.70 trillion in Q1, a 27.9% quarter-over-quarter decrease. This was not a random quarterly fluctuation but yet another quantitative annotation of the "continuing trend of slump" since 2025 — trading activity was slowly but surely trending downward, and as the excitement faded, fewer and fewer people were actually standing in front of the trading tables.
Ironically, the leading exchanges remained robust. Binance consumed about 38.7% of that $1.95 trillion, while Bybit maintained around a 10% share; the concentration at the top was not broken, yet it could not prevent the overall decline. The problem clearly did not lie with a single platform, but rather with the sentiment and expectations surrounding the entire asset class: as trading volume continued to shrink, it indicated that short-term speculation and off-market incremental funds were also retreating. Even if prices did not collapse immediately, they would be more prone to a slow decline or volatile oscillation in a lack of supporting liquidity. This round of retreat was essentially a systematic contraction driven by cooling risk appetite.
Ban on Chinese AI Controversy: Zuckerberg Bets on Openness
At the very moment the market confirmed the systematic cooling of crypto trading volume, the disputes in the AI sector were heating up. On July 29, 2026, Meta CEO Mark Zuckerberg publicly stated in a media interview that "banning the use of advanced Chinese AI models in the U.S. is not an effective solution," directly addressing the internal discussions in the U.S. government about the possibility of restricting the use of Chinese AI models. This was not a technical roadmap, but a political statement: what he opposed was the impulse to respond to competitive pressure with comprehensive bans, which was viewed among U.S. tech giants as a clear warning against excessive technological restrictions.
Zuckerberg's positioning exposed the fractures within the U.S. AI regulatory pathways. On one end are policymakers emphasizing security and geopolitical rivalry, inclined to mitigate potential risks by restricting model sources; on the other end are tech company executives betting on global collaboration and market scale, who fear that restrictions will lead not to "safer AI" but to artificially fragmented markets and forced innovation cycles. For investors who pay attention to both crypto and AI, this is not just a battle of public opinion: the future choice between openness and closure will directly reshape the flow patterns of cross-border data and computational power, thereby influencing the valuation framework and globalization limits of related assets for a long time.
OpenAI Cuts Prices by 25%: Voice Entry War Intensifies
While regulators were still debating whether to "close the door" on Chinese models, OpenAI chose to keep the door wide open: launching two voice transcription models, GPT Transcribe and GPT Live Transcribe, and overall cutting related prices by about 25%. This was not a simple product update but a strategic landing around the "voice entry" — whoever can dominate developers' attention with cheaper and more stable transcription capabilities stands a better chance of making voice the default entry for their ecosystem. Under the pressure from Google and Amazon’s years of deep forays into voice and cloud AI, OpenAI leveraged its price cuts and new features to shift the battlefield from parameters and computing power to the minutiae of "cost per voice minute."
With the cost curve pushed downward, many previously only theoretical AI native scenarios could now start to make financial sense: transcription of entire meetings, real-time quality checks for customer service calls, and data accumulation from offline store conversations — all transforming from "expensive experiments" into "sustainable expenses," leading to deeper penetration of AI into internet products and corporate processes. In stark contrast, during the same period, the spot trading volume of the top ten CEXs shrank from $2.70 trillion in Q1 to $1.95 trillion in Q2 2026, down 27.9%, with liquidity retreating, while AI basic services were employing "price wars and scale" to amplify their imagined future cash flows: attention from funds and entrepreneurs was shifting from highly volatile on-chain markets to these seemingly boring but steadily amplifying underlying capabilities.
Chip Stocks Drop Again: Concerns Amidst AI Power Frenzy
Just as OpenAI’s price cuts and model iterations were making headlines with "accelerating investments in computing infrastructure," the secondary market delivered a starkly different response. On July 29, 2026, pre-market in the U.S., the semiconductor sector extended its downward trend: SK Hynix dropped about 1.9%, Micron Technology about 1.6%, SanDisk about 2.3%, and Nvidia also fell slightly by about 0.3%. These companies, deeply embedded in the AI memory, storage, and GPU supply chains, had long been seen as bellwethers for computing demand, yet today they collectively weakened, amplifying the sense of dissonance where "the industry is soaring, while stock prices hesitate."
Market analysis attributed the pressure to micro-adjustments in expectations for AI chip demand and a return of macro uncertainty, but there was no clear, unified explanation to placate the capital markets. Fundraising and production expansion narratives still ran smoothly in the primary market, with OpenAI pushing for application scale expansion through price cuts, yet in the secondary market, capital began to seriously question: how long can this AI cycle last, how long until the massive investments in data centers and chips pay off, and if additional production capacity comes online, will there be a dual backlash from demand fluctuations and price wars in the coming years? The few percentage points of decline in pre-market trading were not fatal but served as a reminder — the investment pace in computing infrastructure had shifted from "growth only" to a delicate game restrained by stock prices and cash flow expectations.
OceanBase Rumored Financing: Battle for AI Data Foundations
On the same day that Wall Street began tightening its nerves over the computing recovery cycle, the domestic market offered a different answer. Media reported on July 29 that OceanBase, a database company under Ant Group, was gearing up for its first round of external financing, with rumored amounts between 2 billion to 3 billion RMB. The company responded without confirming the specific financing scale or investors, but threw out a more critical line: they were fully dedicated to "data technology and product innovation for the AI era" and would continue to maintain open communication with the capital market. This statement, more than any valuation figure, serves as a signpost for capital — database vendors are no longer just providers of "data storage and retrieval" but are set to reconstitute themselves as the data foundations behind large models.
If chips and racks are the most visible "steel and concrete" of this round of AI competition, then the databases and data infrastructure represented by OceanBase are being repackaged as indispensable "foundations." As global semiconductor stocks face pressure due to fluctuating expectations and OpenAI trades price for scale at the same time, a local Chinese data infrastructure software company chose to go public with capital markets, which was widely interpreted as Asian funds preparing in advance for AI infrastructure: on one side, computing investments are entering a stage of fine-tuned games, while on the other, database-centric vendors are vying to secure strategic positions in data infrastructure. The true determinants of victory are shifting from "whose model is stronger" to "who can master the most fundamental layer of data foundations in the AI era."
Displaced Cycle of Cryptocurrency and AI: Where Will the Money Flow?
The spot transaction of the top ten CEXs in Q2 2026 dropped from $2.70 trillion in Q1 to $1.95 trillion, a quarter-over-quarter decrease of 27.9%, contrasting sharply with the price cuts, regulatory disputes, and infrastructure expansions surrounding AI that emerged intensively on July 29: on one side, cryptocurrency trading activity continues to retreat, with liquidity willingness noticeably cooling; on the other side, Zuckerberg openly opposes banning Chinese models, OpenAI cuts voice service prices by about 25%, semiconductor stocks adjust amidst fluctuating expectations, and OceanBase doubles down on data foundations, outlining an AI capital curve that is still rapidly iterating but has entered a phase of cost and expectation repricing internally. In the short term, cryptocurrency appears more like it has entered a cooling trading period, where prices cannot easily be ignited by sentiment alone; AI, meanwhile, is in a mixed segment of business model validation intertwined with cost reassessment, and the paths of regulation, computing and data infrastructure returns remain undecided, with risks and opportunities highly intertwined. What will truly determine the next round of capital flow will be whether the U.S. implements restrictions on Chinese AI models, whether computing and database foundational facilities can yield sustainable cash flows, and whether cryptocurrency can find new growth points in conjunction with the AI narrative or practical applications. For investors, this time resembles a crossroads that requires "disassembly and observation": keeping a close eye on how policy battles reshape the AI landscape, tracking the investment and return curves of computing and data infrastructure, while also monitoring when CEX transactions and on-chain activity stabilize and recover, instead of being swayed by a single price curve or short-term hype.
Join our community to discuss together and become stronger!
Exclusive Hyperliquid benefits for AiCoin: https://app.hyperliquid.xyz/join/AICOIN88
Exclusive Aster benefits for AiCoin: https://www.asterdex.com/zh-CN/referral/9C50e2
On-chain Telegram community: https://t.me/AiCoinWhaleData
On-chain community: https://www.aicoin.com/link/chat?cid=N6OVMor5g
AiCoin on-chain Twitter: https://x.com/aicoinwhaledata
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
