🦾 Why Data, Not Models, Is the Real Moat in Embodied AI The timing of this question is hard to miss. This week, the World Humanoid Robot Games released a 2,500+ hour dataset covering 12 scenario categories, 44 operations, and more than 10,000 tasks. Crucially, it also includes failures and edge cases. But raw hours tell only part of the story. Zhihu contributor 于超, an assistant professor at Tsinghua Shenzhen International Graduate School, shares his team’s view on why data has become embodied AI’s hardest-to-replicate advantage. 1️⃣ Robot scaling is fundamentally asymmetric Like language models, robot policies appear to benefit from more data, larger models, and greater compute. But these three inputs do not scale equally. Model architectures can be studied and reproduced quickly. General-purpose compute can, in principle, be purchased. High-quality robot data is different. It must be accumulated through physical interaction, and competitors cannot recreate it overnight. That asymmetry is what turns data into a moat. 2️⃣ Robot data must be manufactured LLMs inherited decades of internet data. Robots did not. Every useful trajectory must be produced through physical interaction. Even a simple cup-grasping task changes with the object, lighting, environment, camera angle, and robot body. So raw hours are not enough. What matters is the diversity of embodiments, tasks, objects, failures, and recoveries. Open X-Embodiment needed more than 20 institutions and 22 robot platforms to collect over one million trajectories. DROID used 50 collectors for a year, producing only 350 hours of data. New methods such as UMI and egocentric recording make collection easier. But every hour still requires real people, equipment, and time. 3️⃣ A successful trajectory can still be bad data Robot data quality is more complicated than whether a task was completed. On the hardware side, camera accuracy, encoder readings, force sensors, calibration, communication latency, and synchronization across modalities can all corrupt a trajectory. The human operator adds another source of noise. Teleoperating a robot is not the same as performing the action directly. Operators hesitate, pause, readjust, and develop habits for compensating for the control system. A task may succeed even when parts of the demonstration should never be imitated. Success is therefore only the coarsest possible label. One trajectory can contain both excellent behavior and inefficient or misleading actions. Training on the entire trajectory without distinction effectively tells the robot to learn both. 4️⃣ The next challenge is information density Collecting more trajectories is only half the problem. Teams must also identify which parts are worth learning from. Yu Chao’s team developed STEAM to detect local progress within a trajectory without frame-by-frame annotation or manually designed rewards. It separates useful progress from hesitation, failure, and recovery. The key question is shifting from “How many trajectories do we have?” to “How much useful information does each trajectory contain?” 5️⃣ Embodied data is physically expensive Text can be copied. Videos can be downloaded. Robot data requires a physical production process. Collecting one hour may involve a robot, sensors, teleoperation equipment, an operator, a suitable environment, task materials, and engineers who maintain and calibrate the system. Real factories, stores, and homes add even more complexity. And pressing the record button is only the beginning. Transmission, cleaning, governance, and storage can cost more than collection itself. The author offers a rough calculation. If a company wants one million hours of real-world data and reduces the combined collection and management cost to RMB 200 per hour, the total still reaches RMB 200 million. 🔑 The real moat compounds over time Quantity, quality, and cost explain why embodied AI data cannot be replicated through a short burst of spending. Large, diverse, high-quality datasets require physical infrastructure, operational discipline, and years of accumulation. As robot policies continue to benefit from scaling, the durable advantage will belong to teams that can repeatedly: 🔹 Collect broader real-world experience 🔹 Identify the most informative behavior 🔹 Preserve failures and recovery signals 🔹 Turn noisy trajectories into useful learning data In embodied AI, having data and knowing how to use it are becoming two very different capabilities. And the second may ultimately matter even more than the first. 🔗 Full analysis: #EmbodiedAI# #Robotics# #PhysicalAI# #RobotLearning# #AIData# #ScalingLaw# #Tsinghua#

 美股盘前情报 | 2026年6月3日 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 来源：CNBC · Benzinga · StockTwits · The Motley Fool · TheStreet Pro · Reuters · Bloomberg · LSEG · FTSE Russell · · Yahoo Finance · The Globe and Mail 数据窗口：过去24小时（优先过去12小时） ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【市场快照】 • S&P 500 期货：-0.3% | 纳斯达克期货：+0.4% • VIX：15.77（一年新低，中性偏低） • 恐惧与贪婪指数：66（贪婪区间） • WTI原油：约93美元/桶（伊朗紧张局势提供支撑） • 美元指数：DXY 107.2 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【第一部分：头条新闻 — 科技/AI/半导体】  黄仁勋台北GTC（2026年6月2-3日）：最大单一催化剂 ━━━━━━━━━━━━━━━━━━━━ 来源：The Motley Fool · StockTwits · Bloomberg · Reuters · Yahoo Finance 等 1. MRVL +32.5%（周二）→ 创历史新高。黄仁勋在台上表示：“Matt（Murrett）正在打造下一家万亿美元公司。”MRVL今日盘前再涨+22%。 → 非英伟达定制AI芯片 = 新AI阿尔法。顶级科技基金经理Gary Black：“随着焦点转向定制AI ASIC，Broadcom和Marvell是最大赢家。” 2. AVGO Broadcom：与MRVL一起创52周新高。定制AI网络/DPU芯片在企业端份额提升。Gary Black称AVGO是“定制芯片转型的大赢家”。 3. HPE Hewlett Packard Enterprise：盘前创52周新高。分析师：“AI驱动的强劲季度——股价值得更高估值。”Aruba/AI基础设施 backlog 强劲。 4. 英伟达Vera Rubin平台：今日在台北GTC宣布进入量产。下代AI数据中心GPU。NVDA盘前小幅下跌0.5%，市场正在消化MRVL/AVGO等竞争对手的狂飙。 5. 台积电ADR 创纪录新高 +2.5%。台北GTC确认与英伟达深化合作。费城半导体指数（SOX）大涨5.9%至历史新高——2026年最大单日涨幅。 6. IPG Photonics、MACOM、Amkor：黄仁勋GTC台北 keynote 后集体暴涨。AI光子学 + 先进封装 = 下一大瓶颈。 7. 特朗普行政令：政府优先获得先进AI模型使用权。StockTwits称这是“利好AI基础设施”信号。微软+英伟达合作：RTX Spark + Vera CPU用于Windows AI笔记本。本周Build 2026大会即将举行。 8. MU Micron：瑞银上调目标价至1625美元（潜在上涨超100%）。HBM内存需求进入结构性超级周期。  警示信号：Michael Burry（《大空头》）警告：AI芯片狂热距离2000年互联网泡沫峰值仅差7%。图表显示需对AI momentum股保持谨慎。散户与机构仓位出现分化。 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【第二部分：散户情绪 — StockTwits/TradingView/WallStreetBets】  过去24小时散户最热议个股： • MRVL：StockTwits热度第一。黄仁勋点名后散户涌入，看多评论主导。“MRVL冲200美元”呼声四起。 • AVGO：定制AI芯片主题带动散户兴趣。“这是2026年的英伟达”情绪升温。 • ASTS：结束两连跌。高管集体买入提振信心。SpaceX IPO传闻带来扰动，但长期逻辑完好，散户整体看多。 • RKLB / LUNR / RDW：受SpaceX IPO猜测拖累下跌。空头认为SpaceX可能蚕食发射需求。 • META：散户视作买入机会。尽管有新订阅、裁员和云计划，股价仍落后Mag 7其他公司。“META相对GOOGL/AMZN仍便宜”。 • TSLA：SpaceX合并传闻导致回调。散户大V认为牛市情景可额外增加4500亿美元估值。 • INTC Intel：今日意外大涨。投资者提问“英特尔为什么涨？”AI PC + 代工 turnaround 叙事回归。 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【第三部分：市场主题 — AI轮动 + 小盘股突破】  AI资金从超大盘轮动至中小盘： • 《The Globe and Mail》：2026年小盘股成为AI最大赢家 • Russell 2000小盘价值股年内跑赢成长股9个百分点 • AI狂热正驱动Russell 2000异动（Benzinga） • IWM、VTWO、URTH均闪现强烈买入信号  Russell 2000指数重构进行中：美国股市总市值达75.6万亿美元，同比增加29%。指数调整带来波动与机会。  板块轮动：能源（油价93美元、伊朗因素）+ AI半导体 + 小盘价值股领涨。防御板块（公用事业、消费必需品）资金流出。  中东局势：油价维持90-95美元区间。美国-伊朗紧张升级，能源股（XOM、CVX）获地缘溢价支撑。 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【第四部分：重点个股 — 权威数据】  MRVL — Marvell Technology 价格：约140+美元（盘前+22%）| 52周新高 催化剂：黄仁勋称“下一家万亿美元公司” Gary Black：定制AI芯片大赢家 Morgan Stanley目标价120+美元 叙事：云端定制ASIC（谷歌TPU、亚马逊Trainium客户）。数据中心定制硅片成为长期趋势。 风险：估值偏高（PS 20倍），短期超买。  AVGO — Broadcom 价格：约220+美元 | 52周新高 催化剂：定制AI网络芯片 + VMware协同 叙事：英伟达之后第二大AI芯片公司。 风险：估值已计入强劲增长。  HPE — Hewlett Packard Enterprise 催化剂：AI驱动的超预期季度，Aruba网络与GreenLake AI服务 backlog强劲。 叙事：AI基础设施 + 边缘计算 + 混合云 = 多年增长。  TSMC 价格：446.69美元（+2.5%，纪录新高） 催化剂：与英伟达深化合作，先进制程产能成为结构性护城河。  IPG Photonics / MACOM / Amkor：GTC keynote后集体暴涨。AI光子学 + 先进封装成为新瓶颈。 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【第五部分：美股小盘阿尔法  】  IWM — Russell 2000小盘ETF 触发：小盘价值跑赢9个百分点，AI轮动确认。 催化剂：今日ADP就业数据、本周联储讲话、6月底Russell重构完成。  FLNC — Fluence Energy（AI数据中心冷却） 触发：黄仁勋强调AI数据中心基础设施，液冷/电源成下一瓶颈。 验证：GTC后股价强势上涨，机构买入量增加。  SOXL — 半导体3倍做多ETF 触发：SOX指数+5.9%创历史新高，MRVL+32%。 风险：3倍杠杆存在衰减，仅适合短期战术操作。 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【第六部分：宏观 + 市场结构】  今日重点事件（2026年6月3日）： • ADP私营部门就业（5月）预期+18万 • ISM服务业PMI • EIA原油库存 • 联储官员讲话  主要风险： 1. 伊朗/中东升级 → 油价突破95美元 → 通胀风险 → 联储偏鹰 2. AI芯片泡沫（Burry警告）→ 板块可能出现5-10%回调 3. SpaceX/OpenAI等IPO可能抽走市场流动性 4. 对60个经济体的关税压力 5. 台海紧张导致供应链中断 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【第七部分：机构资金流向】 • AI/基础设施：MRVL、AVGO、HPE、TSM —— 机构大量买入 • 能源：XOM、CVX —— 地缘溢价流入 • 小盘价值：IWM、VTWO —— 2026年首次显著机构轮动 • 流出：公用事业、消费必需品、REITs ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 【投资总结 — 三种情景】  牛市情景：定制AI主题延续 → MRVL/AVGO/TSM领涨 → 小盘爆发 → IWM突破250+ 催化剂：MU、AMD财报超预期 + ADP就业数据温和  熊市情景：Burry警告兑现 → AI芯片见顶 → MRVL/AVGO回调 → 纳斯达克回调3% 催化剂：ADP强劲 + 通胀升温 → 降息预期推迟  基准情景：AI基础设施长期牛市 → 半导体高位震荡 → 小盘分批轮动 → VIX维持15-18 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━  免责声明：本内容仅供参考，不构成任何投资建议。数据来源于公开英文财经媒体。历史表现不代表未来结果。投资前请咨询持牌财务顾问。

22 billionaires are spending $229M against CA's 5% billionaire wealth tax, led by: Sergey Brin worth $273 billion Peter Thiel worth $28 billion Chris Larsen worth $13.7 billion Disgusting. They'd rather let people die because they can’t afford healthcare than pay their fair share.