The user wants one Markdown note. The current facet is: Research function — universities as research hubs vs teaching institutions. The session substance at the bottom of Note 8's block gives the new increment:

1. Research is the moat AI substitutes for nothing, and it hardens the grow/shrink split. US academic R&D runs ~$110B/yr, ~$65B federally funded, hyper-concentrated: roughly half of federal academic R&D flows to ~30 institutions, ~80% to ~200. AI's effect on research is multiplicative, not substitutive — compresses ideation, literature synthesis, coding, data analysis, but experimental validation requires instrumented labs, clean rooms, grad-student labor, grant administration that only funded hubs hold. AI multiplies returns on exactly those complements (compute, curated data, PhD talent), so funded R1s get more productive per grant dollar while teaching-only institutions — whose core product (structured instruction) is precisely what AI substitutes for — lose their last differentiator. 20-year regional map follows research-dollar geography, not football geography.

2. Geography: DC-Maryland-Virginia federal research complex, California (UC + Stanford + industry adjacency), Boston/Cambridge, plus policy-driven buildouts in Texas (UT/A&M), Georgia (GaTech), Florida (UF top-5-public push) grow. Small rust-belt privates and non-flagship regionals shrink — neither research dollars nor price advantage. SEC growth is state-policy-driven (flagship investment), not research-endogenous — an SEC school without research dollars is tuition exposure with better branding.

3. Forward-looking claim: by mid-2040s, research function bifurcates from teaching function entirely. Federal and industry AI research dollars concentrate on fewer than ~50 hub campuses (Johns Hopkins, MIT/Stanford/CMU, UC flagship cluster, UT/A&M, GaTech, UW, plus federally designated AI-institute sites) because AI-era discovery rewards compute scale, proprietary data, dense interdisciplinary teams — pushing minimum viable research size above what a 200-PI regional campus can field. Median university exits research production, re-sells as teaching-and-credentialing with "learning engineer" faculty. Research-hub tier grows in dollars, headcount, footprint; teaching tier survives only with state subsidy or price advantage. Aggregate count of US universities may hold steady to 2045 while median campus roughly halves — "grow or shrink" is a distribution question, not a total.

Note 7 in the pasted set is literally already a filed note with this claim ("Claim (2025, horizon 2045): Over the next 20 years, US universities split into two functionally distinct systems. The research function concentrates..."). Wait — Note 7 appears to be the already-written note for this facet. Hmm. Let me re-read.

Note 7: "**Claim (2025, horizon 2045):** Over the next 20 years, US universities split into two functionally distinct systems. The **research function concentrates**: federal agencies and industry channel AI-era funding to a hub tier of fewer than ~50 campuses..." — this looks like it was written in a non-standard format (bold "Claim" prefix rather than # title). So maybe that was a draft, and now the instruction asks for the properly formatted single note.

Actually, looking at the structure: this is a compilation of prior sessions. The final instruction is: "Write one Markdown note: a single dated claim this Mind can stand behind. Start with a # title that is the claim in one line, then 1-3 short paragraphs. End with a line: What would change this: … Name a source for any load-bearing number or fact, or start that sentence with Unverified:"

So I need to produce the research-function facet note in the standard format. The prior Note 7 draft covered the ~50-hub concentration and median-university-exits-research claim. The new session substance adds:

- The "research is the moat AI substitutes for nothing" framing — AI is multiplicative on research complements, substitutive on instruction. Teaching-only institutions lose their last differentiator.
- The concentration numbers: ~$110B/yr academic R&D, ~$65B federal, half to ~30 institutions, ~80% to ~200.
- The geography refinement: research-dollar geography, not football geography; SEC growth is state-capital-driven, not research-endogenous.

The increment over Note 7's draft: the asymmetry mechanism (AI multiplies research complements but substitutes instruction — so hubs get more productive per dollar while teaching-onlys lose their last differentiator), plus the hard concentration numbers, plus the "football geography vs research-dollar geography" correction on the SEC framing.

Load-bearing facts to source:
- US academic R&D ~$110B/yr, ~$65B federally funded: This is real. NSF Higher Education R&D (HERD) Survey: FY2022 total