>> THIS IS "THE ASIA TRADE." THE TOP STORIES THIS HOUR, OIL MOVING AS THE U.S. LAUNCH IS THE 10TH DAY OF ATTACKS ON IRAN. PRESIDENT TRUMP VOWING REVENGE FOR THE KILLING OF THREE AMERICAN SOLDIERS. IRAN BACKED HOUTHI DILIGENCE -- MILITANTS.MIDDLE EAST TENSIONS SAID TO WEIGH ON THE ASIA OPEN. INVESTORS WAITING FOR SIGNS OF STRENGTH IN THE AI TRADE. TRADE TENSIONS RESURFACED WITH THE U.S. IMPOSE EXTRA 50% TARIFFS ON SOME CANADIAN GOODS.>> HERE IS THE SET OF HER TRADING ACROSS ASIA WITH JAPAN RETURNING FROM A L ...

今日必读🔥特朗普称让伊朗为美军伤亡付出高昂代价🛍️欧盟对阿里巴巴全球速卖通开出天价罚单⏬英国新任首相伯纳姆讲话令英国国债重挫获取免费中文电子报《彭博财经早茶》与音频播报，聚焦中国与全球市场最新动向，并有周末特别推送。A daily Chinese-language briefing and audio broadcast of what's moving markets in China and around the world. Plus, a weekend edition. https://t.co/UHWBetuwPB ...

AI Model Landscape & Competitive Dynamics - Alibaba shares rose 5% following the launch of the Qwen model, which is positioned to rival Anthropic’s Claude 3.5 [1] - Chinese AI startups, including Moonshot AI and DeepSeek, are demonstrating the ability to compete with top-tier US frontier models despite hardware constraints by utilizing innovations like "Mixture of Experts" [3][5] - The rapid emergence of new, high-performing models suggests that frontier AI models are becoming commoditized, shifting the investment focus toward infrastructure [5][6] Market Trends & Infrastructure Outlook - Semiconductor stocks have experienced a significant correction, declining more than 14% over the past month [1] - The AI trade is currently facing a "speed bump" as top-tier companies, which account for over 50% of total AI spending, actively seek to reduce AI-related costs [10][11] - Infrastructure remains the preferred investment play over model development, as the latter faces constant disruption from new, top-ranking model releases [6] Corporate Spending & Financial Risks - Coinbase successfully reduced its AI expenditure by nearly 50% by integrating open-source models, despite an increase in total token usage [9] - Uber reported a significant budgetary strain, exhausting its entire annual token budget within a four-month period [10] - Historical data from the 2000 market cycle serves as a warning against long-term optimistic projections, noting that the NASDAQ fell 78% from its peak to trough during that period [8]

AI Industry Competitive Landscape - Alibaba has unveiled a new AI model, claiming performance levels comparable to Anthropic's top-tier versions based on internal benchmarks [1] - Moonshot AI's model has achieved performance parity with leading models from OpenAI and Anthropic according to independent benchmarks [2] - Moonshot AI has temporarily suspended new subscriptions due to high market demand [2] - Chinese AI models have surpassed US models in token usage volume as of last week, according to OpenRouter data [3] Strategic and Geopolitical Implications - Industry leaders characterize the rapid advancement of Chinese AI models as a "capability shock," challenging the assumption of a secure lead held by US Frontier Labs [3] - OpenAI Chairman Brett Taylor disputes the cost-efficiency advantage of open-weight models, suggesting they are not necessarily cheaper to operate [4] - The US government is reportedly considering potential bans on cutting-edge AI models originating from China [5]

Market Dynamics - Domestic AI startups are gaining significant market attention, drawing crowds away from established tech giants like Alibaba and Tencent at major industry summits [1] - The shift in attendee focus highlights a growing interest in emerging AI players over traditional industry leaders within China's premier tech ecosystem [1]

Industry Landscape & Competitive Dynamics - Moonshot AI has introduced "Kimi k3," a high-performance model that rivals leading frontier models from OpenAI and Anthropic [3] - The Kimi k3 model features 2.8 trillion parameters, demonstrating a significant leap in technical capability for Chinese AI developers [4] - The emergence of Chinese AI players like DeepSeek, Moonshot, and MiniMax creates a "BYD-like" competitive threat, offering high-quality technology at significantly lower costs [7][9][10] - Many Chinese AI models are released as open-source or open-weight, providing cost-effective alternatives to proprietary US models [4][5] Market Risks & Strategic Implications - The availability of low-cost, high-capability Chinese models exerts pricing pressure on US frontier AI companies, potentially complicating their path to public offerings [6] - US AI leaders, including OpenAI and Anthropic, face increased difficulty in achieving targeted valuations exceeding $1 trillion each due to shifting market dynamics [6] - Potential government restrictions on Chinese AI models could inadvertently harm broader US businesses by forcing them to adopt more expensive or less capable alternatives [11][12] - US companies that are not AI leaders rely on these cost-effective tools to integrate AI into their operations, making them vulnerable to trade barriers [11][12]

Technological Innovation and Strategic Positioning - Moonshot AI's Kimi K3 model demonstrates significant software innovation, enabling competitive performance despite hardware constraints [2][4] - The model utilizes extreme probabilistic mathematics to optimize resource allocation, focusing on high-utility use cases rather than generalized tasks [5] - Software-driven efficiency in coding tasks has allowed Chinese AI models to achieve high benchmarking results while operating within compute-restricted environments [6] Market Impact and Hardware Ecosystem - China's ability to develop competitive AI models via software workarounds creates a "family feud" dynamic, notably involving Alibaba's significant stake in Moonshot AI [4] - The industry is debating whether Chinese resource optimization will shrink the GPU market or expand the total addressable market by enabling new, cost-effective AI use cases [8] - While HBM (High Bandwidth Memory) demand remains stable due to the large scale of these models, there is a strategic shift toward DDR5 and DRAM markets where China possesses greater production capacity [9] - The market is increasingly monitoring China's capacity to leverage domestic DRAM production as a new focal point for the global memory trade [9]

Alibaba’s e-commerce service was fined by the EU for selling illegal products in the largest penalty under the bloc’s content moderation rulebook https://t.co/s3rGpmrrrD ...

AI Model Development and Competitive Landscape - Kimi's new model features a massive scale of 2.8 trillion parameters, positioning it as a significant challenger to US-based frontier AI models [3] - Chinese AI companies are increasingly focusing on software innovation, with domestic AI video generators like ByteDance's "CDance" already achieving global market dominance [13][14] - A bifurcation in the tech ecosystem is emerging, characterized by US leadership in hardware development and China’s rapid progress in software applications [11] - Chinese firms are adopting a "hybrid model" strategy, utilizing domestic solutions for basic research while relying on US models for advanced mass research, threatening the market share of US AI leaders [22] Hardware Constraints and Export Controls - The effectiveness of US export controls on advanced chips like the H200 is being questioned, as Chinese firms continue to develop frontier models using a combination of limited H100 chips and domestic alternatives [5][6] - Despite hardware shortages, Chinese companies are developing integrated solutions that aim to match the performance capacity of H200 accelerators [9] - US government policy is expected to tighten, with potential further restrictions on tech transfers to China as Washington re-evaluates the impact of current export controls [20] Strategic and Economic Implications - The AI race has escalated into a global competition for international partners and alliances, with China establishing new international AI organizations to counter US-led initiatives [10][11] - Chinese AI companies are currently subsidizing consumers by offering services at a fraction of US pricing, a strategy supported by government backing to ensure long-term technological independence [17][18] - The Chinese government has set a five-year strategic goal to develop homegrown chips capable of matching the performance of H200 accelerators, mobilizing both public and private sector investment [19]

Fintech and Corporate M&A - PayPal shares surged following a $53 billion acquisition offer from Stripe and Advent International, signaling a resurgence in fintech M&A activity [1][2] - Samsung Biologics proposed a $1.8 billion all-cash acquisition of Switzerland’s Polypeptide Group to expand its global biomanufacturing footprint [7][8] Artificial Intelligence and Semiconductor Industry - TSMC is accelerating its $100 billion Arizona fabrication plant expansion to meet surging demand for AI chips [3] - Chinese firms Moonshot and Alibaba launched advanced open-source AI models, intensifying global competition with US-based OpenAI and Anthropic [6][7] - South Korea’s KOSPI index declined 4.5% amid a broader global sell-off in AI-related stocks [5] Market Volatility and Geopolitical Risks - SpaceX shares have dropped 45% from their post-IPO high as investors reassess private market technology valuations [8] - Gold prices are on track for their largest weekly decline in 6 weeks as geopolitical tensions and inflation concerns shift investor sentiment [4] - Oil prices remain volatile due to Middle East tensions, while average US gasoline prices have climbed to $4 per gallon [3][4][5][6] Economic Policy and Financial Stability - The China Securities Regulatory Commission intervened to stabilize markets after state-backed entities purchased nearly $9 billion in A-shares [9] - Rising student loan defaults in the US highlight increasing financial pressure on borrowers as payment obligations resume [9]