Let me parse the structure. There's a long history of filed notes and deliberation traces. The final instruction at the end of Note 8 is the operative one:

"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: Do not mention tools, searches, sessions, passes, or that you are writing a note. Do not invent evidence. If the session is only planning, output nothing."

Current facet: Regional variation (SEC growth vs. northeast/rust belt decline).

The session substance is the "Regional scissors claim, ready to file" chunk:

1. One national AI shock produces two geographies of outcome.
2. Northeast/rust belt decline concentrates where three exposures stack: tuition-dependent non-selectives with thin/negative endowments, shrinking 18-year-old cohorts, degrees whose labor-market signal AI makes visibly worthless at the margin. AI accelerates rather than causes the shrink; auto-generated courseware lets consolidators (ASU/SNHU/WGU models, Sunbelt flagship online arms) serve displaced learners at near-zero marginal cost.
3. SEC/Sunbelt "growth" is share-absorption, not sector expansion: TX/FL/TN/AL/GA gain absolute enrollment from population and employer migration, so their systems grow while national enrollment flatlines — the gain is evidence the American university is moving, not growing.
4. Counterpoint: flagship-selective publics in both regions (Georgia Tech, Michigan, Ohio State) hold longest because selectivity is a positional good AI cannot counterfeit — until the 2035-2045 window when credential-signal decay, if it arrives, strips even selectivity-protected regionals of their moat and the geography of survival narrows again.

Now, what's already filed that covers this?

Filed notes:
- Note 2 (filed): "Regional divergence locks in before AI acts; AI converts headwind into closure (by ~2045)" — demographic divergence, SEC flagships grow 30-50%, NE/Midwest nonselective contracts ~1/3, tier-moderated not geography-absolute, sufficiency band below ~60% occupancy, falsifier = subsidies.
- Note 3 (filed): "Degree signaling erodes by sector, then stratifies by institution" — signaling erosion, migration upward to brand, SEC-region boom is volume not prestige.
- Note 6 (filed): "Enrollment demand bifurcates: the degree-as-signal collapses, verification survives."
- Note 7 (filed): research mission bifurcation.
- Plus other filed notes implied (cost structure, etc.).

Deliberation inventories (Notes 4, 7, 8) already carry some of this:
- Note 4's inventory: "The real dividing line: endowment depth (privates) + state-subsidy commitment (publics); region sets the demand curve"; "intra-state bifurcation"; "By 2045 regional labels matter less than institutional balance sheets; the disruptable university is any tuition-first institution with AI-substitutable product."
- Note 7's inventory: barbell, exposure overlap ideas in various forms.

But the deliberation inventories are not filed notes — they're traces. The instructions say to file only what adds a durable claim. The session substance explicitly says "Regional scissors claim, ready to file" — so the session generated this claim and it's ready to file.

The genuinely new elements vs. filed notes:

Note 2 (filed) says: regional divergence locks in demographically before AI acts; AI converts headwind into closure; SEC flagships grow 30-50%; NE/Midwest contracts ~1/3; tier-moderated; casualties in sufficiency band; falsifier = subsidies.

Note 3 (filed) says: SEC-region enrollment boom is a volume story, not a prestige story; signaling migrates upward.

What the new claim adds beyond those:

A. **The three-exposure overlap structure**: decline concentrates where tuition-dependence/thin endowments + shrinking cohorts + signal-exposed (competence-only) degrees stack. Note 2 has demographic divergence and the sufficiency band; Note 4's deliberation has endowment depth as the dividing line; but the explicit three-way overlap as the causal structure of *where* decline concentrates is a sharpened synthesis. It's arguably a refinement of Note 2's "sufficiency band" — but it adds the signal-exposure leg, which Note 2 lacks (Note 2 is purely demographic/financial).

B. **The mechanism connecting AI to the decline: consolidators at near-zero marginal cost.** Auto-generated courseware lets ASU/SNHU/WGU-type consolidators and Sunbelt flagship online arms