From Hangzhou to Shenzhen, AI has already turned into a visible performance exam. Alibaba Cloud's answer is very straightforward: for the past 11 quarters, AI-related product revenues have consistently maintained triple-digit year-on-year growth, with the latest quarter ending on March 31, 2026, showing cloud business revenue of about 41.626 billion yuan, a year-on-year increase of about 38%, and external commercialization revenue also maintaining about 40% growth; on the other side, Tencent's founder, Ma Huateng, admitted at the shareholder meeting that a year ago they thought they were "on board," only to find out later that "the boat was leaking," and now they "are still not secure." These giant companies, both on the front lines, submit contrasting reports on AI reflecting entirely different rhythms of investment and return curves. Across the ocean, OpenAI has moved the battlefield to Washington, establishing a policy and lobbying office near the White House named Workshop, with federal lobbying expenditures of about 1 million dollars in the first quarter of 2026; a16z’s federal political donations in the current U.S. midterm election cycle have exceeded 115 million dollars, trying to seize discourse power in the AI regulation game. Meanwhile, Morgan Stanley raised its S&P 500 target from 7800 to 8000, embedding its optimism about AI-driven profits into Wall Street's figures. The stakes in the capital market are not just reflected in valuation stories: on one side is Manic.Trade's SMART Benchmark, which conducts stress tests on AI trading agents using about 387 evaluations, 180 testers, and 46 models; on the other side, monitoring on AiCoin indicates that the address 0xcf6...6eb24 has heavily invested in long contracts related to INTC on the Hyperliquid platform, with unrealized gains of about 1.15 million dollars and total holdings of about 7.65 million dollars. Analysts identified this as one of the top accounts in the INTClong category. These evaluation results and the substantial on-chain bets together form the early signals that a new generation of "AI trading paradigms" is taking shape.

Alibaba Cloud AI Accelerates Monetization with 11 Quarters of High Growth

While retail investors and institutions are still using AI trading agents to game the futures curve, Alibaba has transformed AI into a steadily rising revenue curve. By the quarter ending March 31, 2026, Alibaba Cloud's quarterly revenue was approximately 41.626 billion yuan, a year-on-year increase of about 38%, with AI-related product revenue maintaining triple-digit year-on-year growth for 11 consecutive quarters. This indicates that it is not just a "hot product," but a complete set of infrastructure and business models that are continuously being scaled. Market research agencies generally see Alibaba Cloud as one of the important providers of AI infrastructure in China, emphasizing that it sells not "stories," but computing power and model services—from slicing underlying chip resources to packaging and outputting model as a service (MaaS), with some analyses indicating that this segment’s revenue growth is also very high.

More importantly, this growth does not primarily come from internal digestion. Alibaba Cloud’s external commercialization revenue grew by about 40% year-on-year, indicating that corporate clients are rapidly pushing their businesses onto Alibaba Cloud's AI stack. Alibaba is "feeding" real corporate demand with AI computing power and model services, rather than relying solely on the internal system’s self-circulation to support cloud operations. In contrast, during the same period, players still deemed to need "catching up" in AI foundational capabilities, Alibaba Cloud's AI commercialization appears particularly smooth. This continuous 11 quarters of high growth provide a clear contrast coordinate for Tencent's self-assessment of “the boat is not yet stable” at the shareholder meeting.

Tencent's Catch-Up in AI: Ma Huateng's Leaky Boat

At the Tencent shareholder meeting on May 13, 2026, Ma Huateng's remark that "a year ago, we thought we were on board, but later found that the boat was leaking, and although we are now on board, we still have not stabilized" laid bare Tencent’s mindset change in AI over the past two years: the "boarding" a year ago was more about betting on trends, now realizing that foundational models, computing power, and engineering systems are interlinked, and any conservative approach can turn into a "leaky boat." To stabilize, it is no longer a matter of adding a few AI slogans. Thus, market discourse interprets this statement as Tencent's first public acknowledgment that it is lagging in AI foundational capabilities, and compared to Alibaba Cloud's 11 consecutive quarters of triple-digit growth in AI-related products, this time difference has been amplified into structural anxiety.

The differences are not just in rhetoric but are more manifested in rhythm: Tencent's past investments in self-researched large models and computing infrastructure have been relatively cautious, more accustomed to embedding AI capabilities into social, content, and gaming businesses rather than directly facing enterprise clients' budgets with cloud computing + model services like Alibaba Cloud. As the commercialization of large models entered the practical phase, Tencent began to accelerate its catch-up—strengthening AI-related team and talent layouts, trying to bridge the gap with competitors at the model and application levels, and tying demand for computing power, models, and business lines together through organizational adjustments, using product iterations to hedge against the gaps left by early conservatism. However, whether this boat can be repaired and return to the main channel will depend on whether these adjustments can shorten the already widened time gap with Alibaba Cloud in the coming quarters.

OpenAI and a16z Setting Rules in Washington

While Tencent is still busy catching up, OpenAI has partially moved the battlefield to Washington. In the first quarter of 2026, the company established an office named "Workshop" near the White House, designated as a policy communication and lobbying outpost; during the same period, disclosure from a single public source indicated its federal lobbying expenditure was approximately 1 million dollars. Once only concerned with model parameters and product rhythm, the technology company has begun studying bills and hearing processes seriously, subtly shifting its identity from a "technology provider" to a "rule participant."

Capital has also not missed this redistribution of the landscape. In the current U.S. midterm election cycle, a16z's federal political donations have accumulated to over 115 million dollars, making it one of the largest single donors in this cycle, openly placing "innovation-friendly technology and crypto, AI regulatory environments" within its policy demands. Rather than purely a political donation, this is more like a prepayment for the future AI regulatory framework: as Congress and administrative agencies engage in intense discussions on AI safety, data usage, and responsibility allocation, tech companies and investment institutions are vying for discourse power through offices, lobbying, and checks. The focus of competition in the AI industry is shifting from "whose model is stronger" to "who can secure a seat at the rules table."

Wall Street Bets on AI Market Landing at 8000 Points

As tech companies vie for seats at the regulatory table, Wall Street has already begun pricing this table. Morgan Stanley raised its S&P 500 target from 7800 to 8000 points in the first half of 2026, citing as one of the reasons the belief that AI will lead to sustained productivity gains and drive upward revisions in profit expectations for tech companies. For them, this is not just "200 points" on the index but ties the entire U.S. stock market's valuation anchor to AI-driven profit curves.

More importantly, Morgan Stanley is not alone. Multiple large banks and asset management firms have repeatedly emphasized in their research reports how corporate capital expenditures on AI will translate into revenue and profits, straightforwardly declaring AI as one of the most important growth themes in the current cycle. Some institutional investors have begun increasing their allocations to large tech stocks and AI-themed ETFs. Although the specific holdings and weights have not been disclosed in detail, the direction is already clear: AI is not only seen as a new tool but as the core narrative of a new bull market. In such macro sentiment, all attempts surrounding AI, from enterprise operations to asset pricing, and to new generation trading tools, are more likely to gain tolerance or even premium, providing a background that the market implicitly accepts as "reasonable" for the upcoming AI trading agents and on-chain bets.

SMART Assessment of AI Trading Agents' Gaps

When market sentiment is willing to buy into all "AI stories," Manic.Trade's SMART Benchmark serves as a splash of cold water on the trading desk. It uses multiple evaluations to restore narratives into comparable scorecards: approximately 387 tests, covering about 180 testers and 46 different models, observing the decision performance and stability of AI trading agents in different market environments. The results are not a one-sided victory declaration—some agents outperform the benchmark, while others lag significantly behind, revealing abilities within the same lane in stark contrast on the chart.

The signals provided by this data are direct: current AI trading agents are still in an exploratory phase, and the decisive factors often are not "how large the model is" or "how many parameters it has," but strategy structure, risk control, and the interaction between humans and models. SMART Benchmark's results indicate that small differences in prompt engineering and risk control rule settings among different testers can significantly alter the final curve's direction. For institutions, this kind of assessment provides a verification of "whether it can be incorporated into the risk control system"; for individual traders, it reminds people to view AI as a collaborative tool that requires institutional constraints, not as an entirely unrestricted automatic money-making machine.

On-Chain Address 0xcf6...6eb24 Bets on INTC

If the SMART Benchmark is about "paper simulations," then according to on-chain monitoring by AiCoin, the address 0xcf6...6eb24 has provided a real answer with actual monetary stakes on Hyperliquid. This address has gone long on INTC-related tokens or contracts on the platform, currently showing unrealized gains of about 1.15 million dollars, identified by on-chain analyst @ai_9684xtpa as one of the top accounts in terms of unrealized gains related to INTSC. More crucially, this is just a corner of its entire chessboard—this address has total holdings of about 7.65 million dollars on Hyperliquid, covering not only INTC-related assets but also other crypto tokens and U.S. stock tokens, a typical large-scale, cross-asset, bullish risk appetite account.

After @ai_9684xtpa pointed out this address on social platforms, discussions surrounding its strategy sources and whether it uses AI tools quickly fermented, but there is currently no public evidence to prove that it has connected to a specific AI trading agent. Nevertheless, at the same time that Wall Street views AI as the new driving force for profits, the on-chain funds have chosen to express their views through long positions on INTC, which itself is an extension of the "AI macro narrative": regardless of whether it is high-frequency traders, quantitative teams, or semi-automated scripts behind this, the bet made by 0xcf6...6eb24 encompasses not just the price fluctuations of a single asset, but also the larger question of whether the chip and computing power track can continue to meet expectations in the AI era.

The Next Act of AI and the Capital Market

Alibaba Cloud is pushing itself to the front row of profit transformation with 11 consecutive quarters of triple-digit growth in AI revenue, while Tencent, through "leaky boat" and "not yet stable" at the shareholder meeting, admits to still being in the catch-up phase; these two rhythms together constitute a watershed moment for Chinese tech giants in the AI era. Meanwhile, across the ocean, OpenAI has opened a "Workshop" office in Washington, with lobbying expenditures of about 1 million dollars per quarter, and a16z has politically donated over 115 million dollars in this round of midterm elections, moving the rules war into Congress and the White House; Morgan Stanley's raising of the S&P 500 target to 8000 points directly reflects the bet on the profit cycle amid this technologic and regulatory game. Simultaneously, Manic.Trade’s SMART Benchmark shows, under 46 models and approximately 387 evaluations, that AI trading agents exhibit varied capabilities, still far from being "automatic money-making machines"; according to AiCoin’s data, the address 0xcf6...6eb24 has gone long on INTC-related contracts on Hyperliquid, with unrealized gains of about 1.15 million dollars and total holdings of about 7.65 million dollars, turning this AI narrative into cross-asset attempts on-chain. The next act to watch is whether Alibaba and Tencent can turn AI revenues into sustainable profits, how the U.S. regulatory framework concerning safety, data, and responsibility is implemented, and whether AI trading tools, including those measured by SMART Benchmark and on-chain strategies like 0xcf6...6eb24, can truly navigate through drawdowns and maintain stable performance over a longer period.

Join our community, let's discuss, and grow stronger together!
Official Telegram community: https://t.me/aicoincn
AiCoin Chinese Twitter: https://x.com/AiCoinzh
AiCoin On-Chain: https://aicoin.com/hyperliquid
AiCoin Exclusive Hyperliquid Benefits: https://app.hyperliquid.xyz/join/AICOIN88
AiCoin Exclusive Aster Benefits: https://www.asterdex.com/zh-CN/referral/9C50e2

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。