14 — The Ecosystem & Its Gaps: what's missing, what's needed, what should exist but doesn't

A reframe doc. The earlier docs treat the biohacking landscape as a set of scenes to catalog and roadmaps to score. This one treats it as a dynamic, shapeable ecosystem — flows of money, talent, and attention that could be redirected — and asks the forward-looking question the author actually wants answered: what is missing, what is under-built, what is unconnected, and what should exist but doesn't? It synthesizes the "highest-leverage plug-in" findings already scattered across docs 04/08/10/11/12 and re-casts them as a single Gap Register. It does not re-derive the scene facts or the P-estimates (those live in 00–12, which carry the source URLs and confidence tiers); it builds on them. Web-search-extended and sourced for the funding-flow and gap claims that are new here. DEMONSTRATED vs SPECULATIVE marked; where I'm unsure I say so. Written 2026-05-31.

Reading guide. §1 maps the ecosystem as flows and states the central tension. §2 is the core: the Gap Register (8–12 concrete gaps, each with what's missing / why / what's needed / who could build it). §3 zooms out to system-level needs (funding mechanisms, institutions, talent, evaluation). §4 is the mutability ranking — what a motivated individual/funder can actually move in 2026. §5 is the AGI-conditioning caveat (cross-link the in-progress AGI doc). Ends with a leverage×tractability table and an open-questions list.

TL;DR

1. The single structural fact of this ecosystem: capital, talent, and attention cluster furthest from the stated goals, while the decisive bottlenecks are under-owned. Longevity VC ran ~$6.2–8.5 B in 2024 (Longevity.Technology, ainvest); BCI drew ~$2.3 B in 2024 with Neuralink alone at a $9 B valuation (Sacra, Applying AI); cellular reprogramming absorbed Altos $3 B + Retro ~$1 B + NewLimit >$280 M (doc 02). Meanwhile the field that is the upload path — whole-brain emulation — is <~500 people worldwide (doc 04 §4), brain-preservation pioneer Nectome has raised ~$1 M total (Live Science), and the Brain Preservation Foundation runs on prize money (BPF). The money is loud and far from the goal; the bottlenecks are quiet and close to it. That mismatch is the opportunity.
2. The ecosystem is mostly a set of disconnected silos, not a pipeline. No one owns preservation → connectomics → emulation end-to-end; the wet-bio wing and the digital/uploading wing barely talk; the alignment field and the uploading field treat each other as out of scope despite sharing an end-state. Many of the highest-leverage gaps are not "invent a new thing" — they are connect two things that already exist.
3. The Gap Register (§2) names ~11 concrete gaps. The three highest-leverage: (a) nobody owns the end-to-end preservation→connectomics→emulation pipeline; (b) the cheap, decisive small-organism sufficiency test that would resolve the WBE crux is essentially unfunded; (c) vascularization/perfusion tooling — the shared bottleneck for organs and bodies — is under-funded relative to reprogramming.
4. The most "someone-should-just-build-this" item is the small-organism sufficiency program (§2 Gap B): a funded, sustained effort to take a C. elegans or fly, scan it, translate it, run it, and show it retains a specific learned behavior — the one experiment that would convert "which details are you?" from open philosophy into a tractable science, and it would cost a rounding error on any single longevity round.
5. Mutability (§4): the gaps closest to the goal are also the most shapeable, precisely because they're under-owned — a single funder or a small team can be the field. The inertia-bound parts (reprogramming capital, BCI hardware races, the supplement market) are where individual leverage is lowest despite the dollars. Leverage and tractability point the same direction here, which is unusual and worth exploiting.
6. AGI caveat (§5): conditioning on transformative AI this decade (the in-progress AGI doc) re-weights the register heavily — it grows the preservation, alignment×uploading, and verification gaps and shrinks the slow-bio-science gaps (an ASI would redo connectomics and vascularization faster than any human program). The gaps that survive both worlds are the ones to bet on.

1. The ecosystem as flows, not a catalog

Stop thinking "scenes" (a static taxonomy — doc 00) and think flows: where money, talent, and attention concentrate, and where the leverage actually sits. The two maps don't overlap, and the non-overlap is the whole point of this doc.

1a. Where the flows concentrate (the loud parts)

1b. Where the leverage actually sits (the quiet parts)

1c. The central tension, stated plainly

The dollars and the bottleneck are anti-correlated. Roughly: the reprogramming + BCI + supplement wings have absorbed >$10 B/yr of capital and most of the talent and ~all of the attention, and none of those three is on the critical path to uploading, while two of them (BCI, reprogramming) are routinely sold as if they were. The things that are on the critical path — preservation logistics, connectomics throughput, the sufficiency test, vascularization, alignment×uploading — together command a budget that is a rounding error on a single longevity Series A. This is not a complaint; it's the arbitrage. In an efficient ecosystem the gaps would be priced away; this one is wildly inefficient, which is exactly why a motivated individual or funder has outsized leverage (doc 08 §A.4: the WBE field is small enough to "fit in a single workshop room"). The rest of this doc is a structured list of where that inefficiency is biggest.

2. The Gap Register (the core of the doc)

Each gap: (a) what's missing · (b) why the market/academia under-supplies it · (c) what's needed / what should exist · (d) who could plausibly build it and what it'd take. Gaps are roughly ordered by leverage; the leverage×tractability ranking is §4. Confidence tiers are on the gap diagnosis, not the underlying facts (those trace to the cited docs).

Gap A — No one owns the preservation → connectomics → emulation pipeline end-to-end

• (a) What's missing. The three stages of the only realistic upload path (doc 04 §8 / doc 08 §A) are owned by separate, barely-communicating communities: cryonics/biostasis orgs (preservation), academic connectomics labs + E11 Bio (scanning), and the ~500-person WBE/Carboncopies world (emulation). No org, funder, or roadmap treats them as one pipeline with shared specs. A brain preserved by Alcor/Sparks today is not preserved to a connectomics-imaging spec; the imaging labs work on freshly-fixed mouse tissue, not decades-old human cryopreserved tissue; the emulation theorists work on the fly connectome, not on what a preserved human brain would actually yield. The hand-offs are undefined. C2.
• (b) Why it's under-supplied. No market pulls it together: cryonics is a tiny membership-funded service industry; connectomics is grant-funded basic science with its own (non-upload) goals; emulation is philanthropy/passion-funded. Each optimizes its own metric; nobody is paid to make stage 1's output be stage 2's valid input. It also spans a century of timeline, which no VC and few grantmakers will own.
• (c) What's needed. An end-to-end "preservation-to-emulation" roadmap and standards body — define the preservation quality metric connectomics actually requires (the 2025 practitioner forecast already converged on "synaptic connectivity as the surrogate biomarker," doc 10 §3d — so the spec is half-written), fund a proof-of-pipeline on a small animal (preserve → image the preserved tissue → reconstruct), and publish the interface contracts so each stage can be improved independently against a shared target.
• (d) Who could build it, and what it'd take. A single mission-driven funder (this is a ~$1–10M/yr coordination + grant-making problem, not a $1B one) convening BPF + E11-class imaging + Carboncopies + a biostasis provider. The pieces and people exist (doc 00 lists them); what's missing is the integrator. This is the highest-leverage organizational gap in the whole ecosystem. C2.

Gap B — The cheap, decisive small-organism sufficiency test is essentially unfunded ⭐

• (a) What's missing. The WBE crux (doc 04 §5b, doc 08 §A.1[3]) is "which physical details are functionally necessary to reconstruct a mind?" — an open science problem that gates everything downstream. The decisive experiment is small and well-defined (codex independently proposed the same one, doc 12 ➕): take a C. elegans or a fly, teach it a specific behavior, scan it, translate scan→model, run the model, and test whether the model retains the learned behavior. If yes, you have evidence the captured level of detail is sufficient; if no, you learn what's missing. Almost nobody runs this as a sustained, funded program. C2.
• (b) Why it's under-supplied. It's a negative-result-prone, unglamorous, basic-science question with no product at the end and no aging-reversal press release — invisible to VC and unattractive to grant committees optimizing for publications. The C. elegans cautionary tale (complete wiring map since 1986, still no agreed emulation — doc 04 §2a) shows the field has had the substrate for 40 years and still hasn't run the clean learned-behavior-retention version at scale. It falls in the crack between connectomics (which maps) and neuroscience (which doesn't see it as their question).
• (c) What's needed. A funded, multi-year "sufficiency benchmark" program: pick organisms of escalating complexity (C. elegans → larval fly → adult fly), ground-truth a learned (not innate) behavior, and run the full scan→translate→run→retest loop, publishing exactly which modeled details (weights, signs, neuromodulation, plasticity rules) are necessary for behavior retention. Carboncopies' Brain Emulation Challenge (doc 04 §4) is the seed; it needs real money and a learned-behavior bar.
• (d) Who could build it, and what it'd take. A worm/fly neuro lab + a connectomics group + a modest philanthropic grant (low single-digit $M/yr — trivial against the longevity totals in §1). This is the single most "someone-should-just-build-this" item in the ecosystem: maximal information value per dollar, it converts the field's central open question into a tractable empirical program, and for a fidelity-valuer like the author it answers their actual crux (doc 10 §4c). C2.

Gap C — Vascularization / perfusion tooling is the shared organ/body bottleneck yet under-funded vs reprogramming

• (a) What's missing. A method to vascularize a centimeter-plus construct down to ~10 µm capillary scale that survives host anastomosis (doc 11 §1). cm-scale perfused tissue is DEMONSTRATED (Miller/Stevens Science 2019; Lewis/Wyss SWIFT/co-SWIFT); the macro-to-micro length-scale bridge across a whole organ is unsolved, and it gates both the real bioengineered-organ spine and the organ-sac moonshot.
• (b) Why it's under-supplied (relative to its leverage). Capital prefers the platform/healthspan narrative (reprogramming "age reversal") to the un-sexy plumbing problem; vascularization is a slow, hard tissue-engineering grind with no "reverse aging" headline. Reprogramming pulled $3B+ (Altos alone); vascularization is a handful of academic labs + ARPA-H PRINT. The reprogramming wing isn't even on the organ-supply path (it's a healthspan play, doc 11 §4) — yet it out-raises the actual bottleneck by orders of magnitude. C2.
• (c) What's needed. Multi-modal printing that bridges mm-channel → capillary; host-anastomosis/anti-thrombosis methods; AI for organoid-vascular-fate prediction (doc 11 §6). Plus the adjacent, underrated sibling: organ/tissue preservation & perfusion (normothermic machine perfusion, tissue cryopreservation) — a force-multiplier for xeno/bioengineered/ectopic programs and a bridge to the author's cryonics interest.
• (d) Who could build it. Technical contributor or funder into the ARPA-H PRINT / Wyss channel, or a new focused vascularization shop. Both Claude (doc 11) and the independent codex pass (doc 12) named this the #1 biopunk plug-in because it advances the real pipeline whether or not organ-sacs ever resolve. C2.

Gap D — Brain-preservation quality & logistics (the controllable failure modes) are under-resourced vs the glamorous science

• (a) What's missing. The upload path's only actionable rung today is preservation (doc 12), and its biggest controllable failure modes are logistical, not scientific: ischemic delay between death and perfusion, standby coverage, geographic distance to a provider, execution quality. The 2025 practitioner forecast rated exactly these — geographic delay and poor execution — among the top failure modes (doc 10 §3d, §5a). Yet the ecosystem pours attention into the science (vitrification chemistry debates) and starves the operations (standby teams, transport logistics, regional coverage, QC). C2.
• (b) Why it's under-supplied. Preservation orgs are small and membership-funded; logistics is capital- and labor-intensive with no exit; it lacks the intellectual glamour of the chemistry. And the structural (upload-relevant) branch is even thinner than the biological-revival branch, which at least has standby infrastructure (Alcor/Tomorrow.Bio) — the structural orgs (Nectome ~$1M raised; Sparks small) have the right goal but the least operational muscle.
• (c) What's needed. Operations-grade standby + transport infrastructure for the structural-preservation branch, regional coverage to cut ischemic delay, and published QC metrics (tied to Gap A's connectomics spec). This is an execution problem more than a research one — which makes it unusually tractable.
• (d) Who could build it. A funder/operator willing to do the unglamorous logistics build-out, partnering with existing biostasis providers; or the structural orgs themselves with a capital injection. Closest to personally actionable for the author (live near a provider, arrange fast standby — doc 10 §5a). C2.

Gap E — The alignment × uploading / digital-minds intersection is neglected

• (a) What's missing. Two fields that share an end-state don't talk: AI alignment treats uploading as out of scope, and the uploading field treats "the AI piece" as an external dependency (doc 04 §7, doc 10 §1). Nobody owns the questions at the intersection: what does "uploading" even mean in a world with ASI (does an ASI reconstruct you, and is that you?); digital-mind welfare (an emulated mind is a moral patient that can be copied, paused, deleted — Age-of-Em's nightmare, doc 10 §2); what a "good" experience machine for an uploaded mind requires (consent, exit, preference-stability — doc 09 §6). C2.
• (b) Why it's under-supplied. Disciplinary silos + the topic's weirdness tax (it sat at the sci-fi/academia boundary). This is changing: 2025 was the year digital-mind welfare became fundable — Anthropic ran a model-welfare assessment via Eleos AI, hired philosopher Joe Carlsmith, and Coefficient Giving (ex-Open Philanthropy) opened AI-welfare grantmaking (digital-minds 2025 review). But it's nascent and focused on near-term LLM welfare, not the uploading-specific questions. C2.
• (c) What's needed. Explicit work at the uploading × alignment × digital-mind-welfare triple point: what makes an upload survival vs a copy made to suffer; governance for emulated minds; whether a "satisfies-without-manipulating" benevolent-AI equilibrium (the FiO premise) is even coherent (doc 10 §7 flags this as possibly ill-posed). Cross-link /workspace/safety/ and /workspace/cryonics/not-die/.
• (d) Who could build it. The emerging AI-welfare orgs (Eleos, PRISM, CIMC) extended toward uploading; or a funder seeding a small "digital minds & uploading ethics" program. Low cost, currently near-empty, growing. C3.

Gap F — The trans / DIY-bio harm-reduction ethos is disconnected from rigorous tooling/QC that could de-risk it

• (a) What's missing. DIY HRT is the most mature real-world instance of decentralized, biopunk-style body-modification that exists (doc 05 §3, doc 00) — a working, decentralized harm-reduction culture with community HPLC/LC-MS QC. But that ethos is siloed from the rigorous-tooling world: the QC labs are ad-hoc/volunteer, the dosing knowledge is folk-aggregated, and there's no bridge to the analytical-chemistry, pharmacometrics, or regulatory-science communities that could de-risk it at scale. C2.
• (b) Why it's under-supplied. Legal/regulatory grey-zone scares off institutional partners; the medical establishment treats DIY as something to discourage rather than to equip; and the DIY culture's decentralization (its strength) resists the centralization that funded tooling implies.
• (c) What's needed. Harm-reduction-grade open tooling: validated low-cost assay protocols, open dosing/ outcomes registries, supply QC standards — applying the proven DIY-drug harm-reduction model (reagent testing, DanceSafe-style) to DIY endocrinology and beyond. This is the "decentralized body-mod, done with real QC" the ecosystem gestures at but hasn't built.
• (d) Who could build it. Community labs (Genspace, Counter Culture Labs — doc 01) + harm-reduction NGOs + analytical chemists; the author is already embedded in this scene (doc 05/07), making it unusually accessible to them as an ethos/demand-side contribution. C2.

Gap G — Independent verification / anti-hype infrastructure for a field where money concentrates furthest from the goal

• (a) What's missing. Given §1c (capital clusters far from the goal, and is routinely sold as if it were close), the ecosystem badly needs standing independent verification — yet there's almost none. There's no durable institution that does for biohacking what the Carboncopies fly-upload rebuttal (doc 10 §3c) or Charles Brenner's reprogramming skepticism (doc 02) did ad hoc: order-of-magnitude reality checks, registry-vs-press-release audits, claim-grading. The honest signal (disease endpoints, OOM arguments) is drowned by the press-release layer (doc 00 ledger). C2.
• (b) Why it's under-supplied. Debunking has no business model and makes enemies of the funded incumbents; it's a classic public good with no private incentive. Academia rewards novel claims, not adversarial replication.
• (c) What's needed. A standing "claims observatory" — independent, adversarial, sourced grading of the field's load-bearing claims (the maker-breaker / cross-model approach in this very repo, doc 12, but institutionalized): registry checks, OOM audits, DEMONSTRATED-vs-SPECULATIVE ledgers maintained over time.
• (d) Who could build it. A small funded nonprofit or an academic center; cheap (it's analysis, not bench work). This repo's own method (independent cross-model verification + critique pass) is a working prototype of what it'd look like. C2/C3.

Gap H — Bridges between the wet-bio and digital/uploading wings (they barely interact despite a shared end-goal)

• (a) What's missing. The two roadmaps (doc 08) share one ideological substrate — morphological freedom (doc 05/07) and arguably one end-state (a self that outlives its given body), yet the wet-bio wing (organs/regen/longevity) and the digital wing (WBE/preservation/uploading) operate as separate ecosystems with separate conferences, funders, and talent pools. Martine Rothblatt is cited as the throughline precisely because she's the rare exception that spans both (doc 00). The shared problems go unexploited: cryobiology underlies both organ preservation (Gap C) and brain preservation (Gap D); perfusion expertise transfers; the "keep-the-substrate-viable" problem is common. C2.
• (b) Why it's under-supplied. Different timelines (organs = this decade; uploading = this century), different evidence cultures (clinical trials vs connectomics), different funders. They don't see themselves as one field.
• (c) What's needed. Cross-wing convening + shared-tooling identification — explicitly map the shared bottlenecks (cryobiology, perfusion, vascularization, tissue QC) and fund them as dual-use (helps organs AND preservation). The author's own portfolio (cryonics + biopunk + trans/DIY) already is this bridge in miniature (doc 08 Synthesis).
• (d) Who could build it. A funder/convener who holds both roadmaps at once (rare — most pick a wing); or a shared-infrastructure org around cryobiology/perfusion as the natural meeting point. C2.

Gap I — The gap between longevity capital and the real bottlenecks (capital allocation, not capital amount)

• (a) What's missing. There is no shortage of longevity capital (~$6.2–8.5B/yr); there's a shortage of capital pointed at the bottlenecks. The reprogramming/senolytics wing is well-funded but is a healthspan play, not an organ-supply or uploading play (doc 11 §4). The ecosystem lacks mechanisms that route longevity dollars to the under-owned leverage points (preservation logistics, connectomics tooling, the sufficiency test, vascularization). C2.
• (b) Why it's under-supplied. VC needs venture-scale returns and short horizons; the bottlenecks are public-goods-shaped, slow, or non-commercial (a sufficiency test has no product; preservation logistics is thin-margin; connectomics tooling is a nonprofit). The capital and the leverage are in different financial instruments — and nobody has built the translator.
• (c) What's needed. Non-VC funding vehicles aimed at the bottlenecks: FROs (E11 is the model — doc 10 §3a), prize mechanisms (BPF's model, scaled — the sufficiency test is ideal for a prize), mission-driven philanthropy, advance-market commitments for preservation QC. The longevity billionaires (who have non-VC money) are the natural source; the instrument is missing.
• (d) Who could build it. A longevity funder willing to run an FRO/prize portfolio instead of an equity one; or an intermediary that packages the bottlenecks for that audience. C2.

Gap J — Regulation / clinical-translation pathways for regenerative & decentralized body-mod

• (a) What's missing. Two regulatory gaps. (1) Clinical translation: the organ-supply spine is real and in trials (xeno, miroliverELAP, LyGenesis — doc 02/11) but there's no expedited-yet-rigorous pathway purpose-built for manufactured organs at scale; RMAT exists but preserves the full evidence bar (doc 11 §5). (2) Decentralized body-mod: the actually-existing "deploy outside FDA" path is offshore regulatory arbitrage (Próspera/Minicircle) — evidentially weak and legally precarious (ZEDE repealed, ~$10.6B arbitration pending, doc 11 §5). There is no legitimate middle path between "full FDA" and "lawless offshore." C2.
• (b) Why it's under-supplied. Regulatory design is slow, expert-thin, and politically fraught; the offshore route is a pressure-release valve that substitutes for real reform; "aging isn't an indication" blocks the longevity wing structurally (doc 11 §4).
• (c) What's needed. A defensible, evidence-preserving expedited pathway for regen/manufactured-organ therapies (building on RMAT + the Sep 2025 FDA draft guidance, doc 11 §5) — the realistic version of the bifurcated-2037 report's "Emergency Pathway." And a legitimate harm-reduction framework for decentralized body-mod (ties to Gap F) that isn't lawless-offshore.
• (d) Who could build it. Policy/regulatory-science people (small expert community; ideas matter) + patient advocacy. A genuinely high-leverage non-bench contribution (doc 11 §6). C2/C3.

Gap K — Talent pipeline & legitimacy for the small-but-decisive fields

• (a) What's missing. The bottleneck fields are tiny (WBE <500 people; vascularization a handful of labs; AI-welfare nascent) partly because there's no career path or training pipeline into them and a lingering legitimacy/weirdness tax. A bright graduate student is funneled toward well-funded reprogramming or BCI, not toward connectomics-throughput tooling or the sufficiency test. C2.
• (b) Why it's under-supplied. Talent follows funding and prestige (§1a); the high-leverage fields have neither at scale; "mind uploading" still reads as fringe in many departments (FHI's 2024 closure, doc 07, removed an institutional anchor).
• (c) What's needed. Fellowships, named positions, and legitimacy-building for connectomics tooling, emulation, preservation science, and uploading-adjacent alignment — the Anthropic Fellows / Eleos model (doc context) extended to the upload pipeline. Make the high-leverage fields fundable career paths.
• (d) Who could build it. Funders willing to endow fellowships; universities willing to host; the emerging AI-welfare institutional scaffolding as a template. C3.

3. What's needed at the system level (beyond individual gaps)

Stepping back from specific gaps, four systemic deficits recur across the register:

1. An integrator / pipeline owner. The deepest structural problem (Gaps A, H, I) is that no one holds the whole. The ecosystem is silos optimizing local metrics; the leverage is in the hand-offs nobody owns. The highest-impact organizational innovation would be a body that defines specs across the preservation→emulation pipeline and across the wet/digital wings, and routes capital to the seams.
2. Funding instruments matched to public-goods-shaped bottlenecks. VC can't fund a sufficiency test or preservation logistics or a claims observatory (Gaps B, D, G, I). The ecosystem needs more FROs (E11 model), prizes (BPF model, scaled), and mission philanthropy pointed at the leverage points — and an intermediary that translates abundant longevity capital into those instruments (Gap I).
3. Evaluation & benchmarks. A field "where money concentrates furthest from the goal" (§1c) is one that cannot self-correct without external evaluation. Two benchmark deficits stand out: the sufficiency benchmark (Gap B — the field's central open question has no agreed empirical test) and standing claim-verification (Gap G — no durable anti-hype institution). Benchmarks are cheap and they reshape where talent and money flow.
4. Talent pipelines & legitimacy (Gap K) for the small fields, so that the leverage points stop being "a single workshop room" of people.

What action would most change the trajectory? In order: (i) a funder who builds the integrator (Gap A) and runs a bottleneck-funding portfolio (Gap I) — this single actor could reshape the whole upload wing; (ii) the sufficiency test getting funded (Gap B) — cheapest high-information move in the ecosystem; (iii) operations-grade preservation logistics (Gap D) — the most personally-actionable and the rung every later stage depends on.

4. Mutability — what a motivated individual/funder can actually move in 2026

The reframe the author asked for: treat the landscape as shapeable, and ask which parts actually bend. The striking result is that mutability and leverage point the same way — the under-owned bottlenecks are both the most decisive and the most movable, precisely because they're under-owned. Money doesn't make the loud parts movable; it makes them crowded.

Most shapeable by a motivated individual/funder in 2026: - The sufficiency test (Gap B) — near-empty, cheap, one funded program is the field. Highest mutability. - The preservation→emulation integrator (Gap A) — the people exist, the integrator doesn't; convening is cheap. - Preservation logistics (Gap D) — execution problem; partly personally actionable (proximity, standby). - Claims observatory (Gap G) — pure analysis; this repo is a working prototype. - Alignment×uploading / digital-minds (Gap E) — newly fundable (2025), still near-empty, growing. - DIY-bio QC tooling (Gap F) — author already embedded; community-buildable.

Inertia-bound (least worth an individual's marginal effort despite the dollars): - Reprogramming capital allocation — $3B+ incumbents; one more funder doesn't move it, and it's off-path anyway. - BCI hardware races — well-funded multiplayer ($2.3B/yr); marginal leverage low; off the upload path. - The supplement/wellness market — huge, commercial, self-sustaining; not shapeable toward the goals. - De-novo ectogenesis & full CNS re-fusion — scientifically inertia-bound (categorical walls, doc 11 §2/§3), not money-bound; no individual unlocks these by 2037. - Regulatory reform (Gap J) — shapeable in principle but slow and politically gated; medium-term, not 2026.

Ranked: gaps by tractability × leverage

5. Caveat: conditioning on near-term AGI re-weights the register

The above register implicitly assumes a human-paced future (humans slowly solve connectomics, vascularization, emulation). Conditioning on transformative AI / ASI this decade (~2027–2032) — the question the AGI-conditioned analysis takes up — changes the weights substantially. (That analysis is the AGI-conditioned codex run, in progress at research/codex-independent/agi-conditioned.md; its framing is in codex-independent/_prompt-agi.md. This section flags the re-weighting; defer to that doc for the conditional probabilities.)

The core logic (from doc 10 §2c and the AGI prompt): if aligned ASI arrives this decade, the binding constraint for uploading stops being "can humans solve WBE?" and becomes "(a) is the ASI aligned, and (b) did you physically survive — brain structure intact — to the point ASI can do the scan/emulation for you?" An ASI could plausibly crack connectomics throughput and the "which details" science fast. So the calculus flips from decades of bio-science toward preservation NOW + alignment.

How each gap re-weights under near-term-AGI:

The gaps that survive — even grow — under both conditionings (human-paced and AGI-this-decade) are preservation logistics (D) and alignment × uploading / digital minds (E). Those are the robust bets: they are high-leverage whether or not ASI arrives soon, because under AGI they become the game, and without it they remain the actionable rung and the valence-decider. The slow-bio-science gaps (C, B-partial, K, H) are the ones an AGI-believer should down-weight — not because they're wrong, but because an ASI would redo them. The honest meta-point (cross-link the AGI doc): if you genuinely believe AGI is <8 years out, the ecosystem's center of gravity should shift from "advance the bio-science" to "survive to ASI + make ASI go well" — which re-ranks this register toward D and E and treats much of the rest as either a bridge or work ASI will redo.

6. Top gaps by leverage × tractability (prioritized)

(Under near-term-AGI conditioning, §5: promote D and E to the top two and down-weight C and the tooling half of A/B.)

7. What I couldn't verify / open questions

• Exact "WBE field <500 people" headcount — carried from doc 04 §4 (a 2025 census estimate); I did not independently re-verify the number, only its order of magnitude (the field is unambiguously tiny). C3.
• Whether nobody funds the sufficiency test — I found no funded program running the clean learned-behavior-retention loop, and codex independently flagged the same absence (doc 12), but absence of evidence isn't proof; a small academic effort could exist under the radar. I'm confident it's not a resourced, field-scale program; I can't prove zero. C3.
• Funding-flow figures — longevity VC ~$6.2–8.5B (2024) and BCI ~$2.3B (2024) come from industry/market-research sources that disagree and slice categories differently (the same ±2× problem doc 06 flags for the "biohacking market"); treat them as directional, not precise. The relative gap (billions vs ~$1M for Nectome) is robust to the imprecision; the exact totals are not. C3.
• Whether a single "integrator" org is the right structure (Gap A) vs. distributed coordination — I assert the coordination gap exists (the stages demonstrably don't share specs); the best organizational form to fix it is my judgment, not established. C3.
• The AGI re-weighting (§5) — this is conditional analysis, not forecast; it depends entirely on P(transformative AI by ~2030), which is itself contested (the AGI doc anchors it). The direction of the re-weighting (D and E grow, slow-bio shrinks) is robust given the conditioning; the magnitude depends on the timeline you believe. C3.
• Digital-mind-welfare funding scale — I confirmed it became fundable in 2025 (Coefficient Giving/Eleos) but did not pin a total dollar figure; it is clearly nascent (orders of magnitude below the longevity/BCI flows). C3.
• Whether DIY-bio QC tooling (Gap F) can scale without triggering regulatory crackdown — genuinely open; the same legal grey-zone that under-supplies it could kill a scaled version. C4.

Reframe/synthesis doc — extends the "highest-leverage plug-in" findings in docs 04/08/10/11/12 into an ecosystem-gap map; funding-flow and gap claims web-sourced inline (figures directional per §7); scene facts and P-estimates trace to docs 00–12 and the bifurcated-2037 report, which carry the source URLs and confidence tiers. Treats the landscape as MUTABLE and forward-looking. Cross-links: 08-two-roadmaps.md (the roadmaps), 10-fio-deepdive.md / 11-biopunk-deepdive.md (the plug-in rankings this re-frames), 12-cross-model-reconciliation.md (cross-model leverage convergence), and the AGI-conditioned analysis research/codex-independent/agi-conditioned.md (§5). Not medical, financial, or existential advice.