Jordan Nanos × Doug O'Laughlin × Max Kan × Joey Brookhart

“Pacing” would slow Anthropic’s capability progress without halting training or compute purchases, potentially weakening its strongest internal model.Near-term scarcity and safety workloads keep compute demand elevated, while semiconductor signals increasingly depend on frontier-lab ARR and capacity premiums.Bank hacks, data leaks, or AI-assisted biological attacks could accelerate regulation.

李翔 × 王华东

具身智能正经历中国投资史上前所未有的高估值融资潮，但王华东认为“贵不是问题”，关键在于能否交付有价值、产生毛利的产品。宇树在2021年B轮投前估值约7亿元时，已按出货量位居四足机器人全球第一；行业仍须警惕以更贵、更不可控的方案重做Rule-based已解决的问题，并观察Room-based落地与迭代能否兑现商业化。

泓君 × Mark × 徐子扬

推理竞争正从峰值算力转向每百万Token成本，decode反复读取全模型，使带宽成为核心瓶颈，SRAM的带宽经济性因此显现。Groq依靠编译期调度，Cerebras依靠晶圆级集成，但MoE路由、通信、良率、供应链与动态增长的KV cache仍是规模化验证的关键风险。

Nathan Labenz × Anton Leicht

Anton Leicht now sees misalignment cases as a major future problem, but says a frontier-only six-month pause would surrender America’s decisive chip advantage while China advances elsewhere.A government halt could signal crash risk, whereas a lab-led reliability pause might remove political tail risk; watch executive action, the 2028 primaries, and Taiwan’s unresolved threat to AGI end-games.

Jensen Huang × President Trump

Jensen Huang cites failed forecasts on radiology, AI-generated code and entry-level jobs to challenge doomsday narratives, while arguing that AI incidents so far call for root-cause engineering and independent evaluation rather than sweeping regulation.With 80% of AI-native companies receiving $400 billion in recent venture funding using open models, NVIDIA is building infrastructure bottlenecks as Jensen predicts China will reach advanced lithography by 2030, leaving commercialization, lab controls and data-center execution to monitor.

Myron Xie × Jordan Nanos

Rubin Ultra shifted from the GTC-previewed 1TB of HBM4E to 192GB of 8-high HBM4, below Blackwell Ultra’s and vanilla Rubin’s 288GB, as HBM supply rations TSMC-secured logic.Four-high saturates the interface at far lower cost per bandwidth and can yield roughly twice as many cubes as eight-high, but model-size growth is the key risk while memory tightness is not expected to ease within this decade.

Raymond × Sam Gao 高岱恒

AI人才正在被代表作而非履历重新定价：硅谷总包过亿美元者或仅几十人，Meta可将年总包50万—100万美元的人抬至近5000万。多模态算法边际价值下行、OPD与基础设施上行且技能半衰期或仅18个月；DeepSeek十年vesting和禁写博客暴露留才机制的风险，Dink的PLG与企业招聘定位仍待验证。

Peter Diamandis × Salim Ismail × Dave Blundin × Alexander Wissner-Gross × Emad Mostaque

AI progress is becoming a measurable cost collapse: Astra beat o3’s ARC-AGI-1 result for $20 versus $500,000, while OpenAI claimed a 10,000-agent Navier–Stokes breakthrough.Data delivered 12× compute-efficiency gains versus 3.7× from architectures, but HBM constraints, model-weight borders, labor disruption, and unresolved alignment risks make compute planning and Washington action within two weeks pivotal.

Jordan Nanos × Eric Wen × Nigel Chiang × Nico Bontigui

Time to power, rather than cost, is driving modular adoption as compute deals reach $40M/MW and some Anthropic configurations exceed $100 million per megawatt.Factory parallelization cuts fit-out from up to nine months to three, shortening builds from 18–24 months to as little as 12; Level 5 commissioning can take 3–8 months, while module lead times reach 18 months.

Nathan Labenz

China’s AI rise looks like a reversion to the historical mean, not an anomaly, while Nathan Labenz argues for “Pax Robotica” rather than victory.Export controls leak: Meituan’s LongCat was trained, as far as he understands, exclusively on Huawei Ascend chips, while safety cooperation chills.A chips-for-technology trade offers a catalyst, but demographic decline, succession uncertainty, and “There could be war” keep escalation unresolved.