Aihedge Leopold Aschenbrenner

Leopold Aschenbrenner AI Trend Research

Use this skill for qualitative-first industry trend research that starts with Leopold Aschenbrenner's public thesis, then translates it into investable AI-infrastructure opportunity maps.

This skill is for research, not financial advice. Always separate facts, sourced claims, inference, and speculation.

Core Thesis Lens

• Treat frontier AI as an industrial mobilization, not only a software cycle.
• Start from the timeline question: how likely is AI researcher-level capability around 2027, and what must be true for that to happen?
• Translate capability scaling into bottlenecks: power, interconnect, data centers, GPUs/accelerators, memory/storage, cooling, gas and grid equipment, transformers, lab security, and sovereign compute.
• Watch for "picks and shovels" where demand is less dependent on one winning model lab.
• Track national-security and US-China competition as catalysts for capex, export controls, domestic supply chains, and cyber/physical security.
• Treat public holdings as clues, not endorsements. 13F filings are delayed, long-only US-listed snapshots and can omit shorts, non-US positions, private investments, derivatives details, and current changes.

Workflow

1. Refresh sources before making current claims.

   • Pull latest SEC 13F filings for Situational Awareness LP / CIK 2045724.
   • Read the latest original interviews, essays, and fund-related public reporting.
   • Use source-map.md as a starting checklist, then browse for newer material.

2. Extract the narrative.

   • Summarize AGI timeline assumptions, compute-scaling assumptions, "unhobbling" assumptions, and security/geopolitical claims.
   • Identify where the thesis has changed or become more/less supported by later evidence.
   • Flag controversial assumptions explicitly.

3. Build the bottleneck map.

   • Convert the narrative into industry nodes: power generation, gas turbines, nuclear/SMR, grid equipment, data-center developers, cloud/neocloud, colocation, fiber/optics, networking, GPUs, ASICs, memory, storage, cooling, security, and defense AI.
   • For each node, state why it benefits, what evidence would confirm it, and what would break the thesis.

4. Map holdings to thesis.

   • Group current and prior 13F positions by theme.
   • For each holding, infer the likely thesis only when the link is supported by filings and public narrative.
   • Distinguish common shares from calls/puts and note notionals/market values when available.

5. Generate stock opportunities.

   • Prefer a watchlist organized by theme and evidence quality rather than a simple buy list.
   • Include direct beneficiaries, second-order suppliers, bottleneck owners, and "if the market is overfitting to one obvious name" alternatives.
   • Add catalysts, valuation sanity checks, liquidity/short-interest risk, balance-sheet risk, and regulatory risk.

6. Produce a research memo.

   • Lead with the qualitative thesis and the highest-conviction bottlenecks.
   • Include source links, current holdings table, opportunity map, risks, and "what to monitor next."
   • End with a clear non-advice disclaimer.

Output Template

Provide:

1. Current source check and dates
2. Aschenbrenner thesis in 5-8 bullets
3. Latest public holdings grouped by theme
4. Industry bottleneck map
5. Stock opportunity watchlist
6. Catalysts and disconfirming evidence
7. Key risks and valuation caveats
8. Final research verdict

Research Discipline

• Use primary sources first: SEC filings, Aschenbrenner's own site, full interviews/transcripts, company filings, and earnings calls.
• Use third-party summaries only for discovery or triangulation.
• Do not claim a current holding without checking the latest filing date.
• Do not treat delayed 13F holdings as real-time portfolio data.
• When recommending further research, specify the exact data to gather next: earnings-call language, power contracts, MW pipeline, capex guidance, backlog, customer concentration, debt maturity, dilution, or regulatory filings.