Gallery

# Conclave Prompt — The China–US Frontier AI Gap *Prepared for Matt Murray · Substrate Capital · July 7, 2026* --- ## Prompt (paste into Conclave) **Is it true that Chinese frontier AI models are now only about six months behind US frontier models, and if so, what actually explains how they closed the gap this fast?** Treat this as a single decidable claim on the gap, followed by a causal account. Do not hedge into "it depends." A verdict of "the question is malformed, here is the better question" is a legitimate outcome and should be argued for on the record if any advocate believes it. ### Definitions the room must accept before arguing - **"Frontier US models"** = the current flagship from OpenAI, Anthropic, and Google DeepMind as of the deliberation date. - **"Frontier Chinese models"** = the current flagship from DeepSeek, Alibaba (Qwen), Moonshot (Kimi), Zhipu (GLM), and ByteDance (Doubao/Seed). - **"Six months behind"** = the elapsed time between a US capability level being first shipped in a generally available model and a Chinese lab shipping an openly available model that matches it on a basket of public evals (MMLU-Pro, GPQA, SWE-bench Verified, LiveCodeBench, AIME, Arena-Hard, long-context retrieval). Cite specific model pairs and release dates. - The room may contest these definitions in Phase 1 but must adopt a shared working definition before Phase 2. ### The gap question — answer with a confidence interval, not a single number Is "~6 months" defensible today, optimistic (gap is smaller), or stale (gap has widened or closed further)? Give a range, and break it out by capability domain — text reasoning, code, math, long-context, multimodal, agentic/tool-use, open-weights leadership. A verdict of "6 months ± 3 months on text reasoning, 12–18 months on agentic workloads, ~0 months on open weights" is more honest than a single number. Anchor to named model pairs, not vibes. ### Structure the causal deliberation in two phases **Phase 1 — Independent nomination (Briefing → Submit stages).** Each of the six advocates must, before seeing the others' submissions, nominate the **two causes they believe are most load-bearing** for the gap-closure. Advocates are explicitly instructed not to converge on a canonical Western-analyst list. Constraints on the nomination round: - At least one advocate must argue the premise itself is wrong — either the gap is not ~6 months, or "gap" is the wrong frame entirely. - At least one advocate must nominate a cause that would not appear in a standard US think-tank readout. Steelman a Chinese-industry, hardware-supply-chain, or benchmarks-skeptic view. - One advocate is designated the **red-team seat** and must argue the strongest version of "the gap is a mirage — Chinese models look close on public benchmarks because those benchmarks are saturated, contaminated, or gameable via post-training, and the real frontier (long-horizon agents, tool use, novel scientific reasoning) still has a 12–24 month gap." **Phase 2 — Collate, contest, rank (Collate → Cross-examine → Debate stages).** Merge the nominated causes into a working set. Before producing any ranking, the room must: - Produce a **one-paragraph causal sketch** identifying which causes are independent drivers vs. downstream effects of other causes. Export controls *cause* algorithmic efficiency; open weights *enable* distillation. A flat ranking hides this — a sketch surfaces it. - Flag any cause that appears in fewer than two advocates' nominations as a **minority hypothesis** and give it a dedicated defender before the room is allowed to dismiss it. - Explicitly name **what is NOT in the working set and why** — one sentence per omitted candidate. At minimum, the room must consider and either adopt or reject on the record: benchmark contamination and eval overfitting; Chinese-language and industrial/government data advantages; weaker IP and copyright friction on training data; US labs d

For AI data centers, the industry is shifting from traditional 415 VAC / 54 VDC power distribution to an 800 VDC architecture, driven by ~1 MW+ rack densities (NVIDIA's "Kyber" / Rubin Ultra generation, the OCP "Mt. Diablo" ±400 V DC effort, and the power ecosystem behind it — Infineon, Navitas, Vertiv, Eaton, Delta, Schneider, TI, et al.). My current understanding of the prevailing assumption is that 800 VDC reaches first volume production deployment around 2027, timed to NVIDIA's Rubin Ultra / Kyber racks, and becomes the default for new large AI builds shortly after. I want you to pressure-test that assumption, not take it as given. Answer in four parts: 1. BASELINE. First, state what the prevailing/consensus rollout timeline for 800 VDC in AI data centers actually is today (correct my characterization above if it's wrong or oversimplified). Cite who is asserting it and what milestone they mean. 2. VERDICT. Is that consensus timeline REALISTIC, OVERLY AGGRESSIVE, or OVERLY CONSERVATIVE? Take a clear position and defend it. 3. YOUR FORECAST. Give the panel's own most-likely (P50) timeframe, with a plausible range, for each of three distinct milestones — answer them separately because they will not land at the same time: - M1 — First production AI data center running 800 VDC end-to-end at rack scale. - M2 — 800 VDC as the default architecture for the majority of new ≥100 MW AI capacity coming online. - M3 — Mature multi-vendor ecosystem (finalized standards, second-source parts, field reliability data). 4. GATING CONSTRAINTS. Name the specific bottlenecks that actually set the pace — e.g. solid-state breakers / DC protection, high-voltage DC-DC conversion (GaN/SiC supply), busbar and connector standards, safety/codes and electrician practices, utility and facility-side readiness. For each, say whether it's a hard blocker or a soft one, and what would have to be true to pull the timeline earlier or later. Where you disagree with each other on the date, make the disagreement explicit rather than splitting the difference.

Who is Tony Chang, and what wealthy, well-connected South Korean family is he a part of? Who is his mother? The only thing I know about Tony Chang is that he works for this company rtwusa.com, and his family is very connected in South Korea.

For a hyperscaler build out, what are the top key build out milestones that would be identifiable from 30cm satellite imagery and support the verification of common publicly reported milestones?

Please review this document and assert the top scientific arguments that are not defensible or are weak, as well as the top three additions we could make to strengthen our scientific argument in support of the overall memo.

The attached memo argues against permitting a hot-mix asphalt plant at the Canyon Rock Quarry site in Forestville, in California's Russian River Valley, using a CEQA-aligned framework. Acting as adversaries to this memo, where does it succeed or fail on two fronts: (1) are its scientific and technical claims defensible enough to survive challenge by the applicant's own air-quality experts and CEQA counsel, and (2) what incremental or additional claims, if any, would most strengthen the case against the facility? For each scientific claim, judge whether it is solid, defensible-but-unsupported, overstated, or rebuttable, and identify which claims help versus hurt the memo's credibility. Then converge on a prioritized verdict: what to cut or soften, what to add or strengthen, and the single highest-leverage change.

The attached memo argues against permitting a hot-mix asphalt plant at the Canyon Rock Quarry site in Forestville, in California's Russian River Valley, using a CEQA-aligned framework. Acting as adversaries to this memo, where does it succeed or fail on two fronts: (1) are its scientific and technical claims defensible enough to survive challenge by the applicant's own air-quality experts and CEQA counsel, and (2) what incremental or additional claims, if any, would most strengthen the case against the facility? For each scientific claim, judge whether it is solid, defensible-but-unsupported, overstated, or rebuttable, and identify which claims help versus hurt the memo's credibility. Then converge on a prioritized verdict: what to cut or soften, what to add or strengthen, and the single highest-leverage change. Include where the panel still disagrees.

What are the best practices for measuring sales cycle days in a SaaS business? Are they different for renewals versus new logos? When does the clock start on the sales cycle?

Does The Third Man ultimately endorse Harry Lime’s cynical worldview or Holly Martins’ romantic moralism?

Who is the real villain of the movie The Third Man, Holly Martins or Harry Lime?

Should I invest in Keyence or Shin-Etsu? Defend your choice.

If i can only pick one and my money will be locked up for 24 months, should I invest in Keyence or Shin-Etsu?