Harry Stebbings × Erik Bernhardsson

Modal has reached $1 billion ARR after 15X growth in 12 months, yet capacity spending keeps cash flow negative.Its 25-neocloud pool, per-GPU-second pricing and adjacent products aim to smooth demand, retain customers and lift gross margins toward Bernhardsson’s 40–60% range.Watch whether 30–40% monthly growth can support buying roughly two to three times projected capacity as hardware-price timing remains unknowable.

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

Anthropic’s proposed $1.5–2 trillion IPO pairs $16.3 billion in first-half 2026 revenue and nearly $1 billion in Q2 operating profit with more than $500 billion in compute commitments.Neocloud declines expose financing risk, with rivals reportedly borrowing at 15% versus SpaceX’s 8%; recursive self-improvement could eventually shock compute demand.Kimi K3.1 Pro’s score of 55, planned open weights and $2/$10 per million input/output token pricing add pressure on closed-model economics as reliability and task completion gain importance.

Gavin Baker

Gavin Baker calls Anthropic’s S-1 an upside surprise: about half the Fortune 500 has adopted it, with encouraging profitability, cash generation and unit economics.Open-weight infrastructure companies outgrow OpenAI and Anthropic as demand broadens, while GPU rental prices rose daily in October and forward capacity remains expensive into 2027.Potential IPOs and higher cloud pricing could ease reliance on external financing for the 2027 build-out, though Baker is unsure whether external financing has peaked.

庄明浩

OpenRouter开源模型占token量71.8%、花费26.9%，平台不代表全网。核心巨头2026年CapEx预测约8000亿美元，高于约7000亿美元自由现金流，当年举债已达2480亿美元。

Harry Stebbings × Naveen Rao

Rao sees US spare power becoming scarce within years, tying AI growth to construction.Unconventional co-designs models and hardware for efficiency as falling token prices pressure inference margins, despite a $10 trillion market forecast within 10 years.He sees robotics 100 times larger than agentic coding, with prototypes in two to three years and a market in five to seven.

Andy Jassy × Ben Horowitz × Erik Torenberg

Amazon expects about $220 billion of 2026 CapEx, yet AWS capacity is sold out for 2026, largely for 2027, and partly for 2028; AWS AI has exceeded a $25 billion revenue run rate.Data centers require investment roughly two years before monetization, pressuring free cash flow; broader enterprise inference adoption also hinges on costs Jassy still calls an order of magnitude too high.

Harry Stebbings

Fund-scale return targets narrow venture: $50–100M seed valuations and $300–500K allocations can make an $18B winner immaterial.CoreWeave must fill capacity and fund buildout; Modal topped $1B run rate after raising $750M at $15.75B, amid unresolved inference competition.Town at $1B and Anduril at $90B highlight entry-price risk, with Anduril’s roughly 40x revenue demanding substantial execution.

Ben Horowitz

CoreWeave’s 953-basis-point CDS implies 16% annual default risk; fixed-price contracts and $35 billion more to raise leave Chamath’s team estimating a $5 billion break-even gap.Sacks sees demand holding, but financing costs remain a risk; Chamath urges Anthropic to raise at least $200 billion for compute, with SemiAnalysis estimating it and OpenAI will take 60% of next year’s global additions.

Harry Stebbings

Pro rata rights emerge as economically valuable options, while Meta’s Muse Wonder signals a return to the model frontier despite unresolved usability questions.Meanwhile, rumored NVIDIA-Reflection and Jev’s $870 million Series A at a $7.5 billion valuation hinge on strategic control, hyper-accurate classification, and whether AI demand can absorb steep inference-price cuts.

Martin Shkreli

Shkreli limits quantum’s edge to factoring and elliptic-curve logarithms; current machines lack quantum RAM.He calls industry spending speculative and expects analog optical computing to challenge Nvidia’s H100, B100, and B200 within 20 years.His pharmaceutical thesis favors mispriced discarded molecules; he calls AI drug discovery within five years “nonsense” and expects no AI-related drug in the next 10 years.