blog: Generally Correct, Specifically Wrong — the grip problem AGI timelines skip

Reframes hallucination as a domain-placement failure, not a data error: general
capability and domain grip are different axes, so recursive self-improvement
without a grip instrument compounds drift instead of curing it. Responds to a
recent long-form AGI-timeline debate (hedged honestly — transcript was not
independently retrievable) betting on next-token-predictor + external-memory
architectures closing the gap to AGI.

Includes a real, unplanned example: our own prompt-time lens misrouted THIS
exact request (write a blog post) into the on-chip Rust walk domain instead of
a content domain, live, mid-composition — the cleanest possible evidence for
the post's own thesis, logged as a routing-reliability gap rather than edited
out.

Originating-Terminal: 🎤 voice
Relevant-Rooms: 🎩 operator, 🧪 laboratory, 🔨 builder
Story: Operator watched a long-form AGI debate and wanted the repo's
decidability/domain-grip thesis applied to it directly — hallucination
reframed via "generally correct, specifically wrong," six needs mapped
explicitly (Connection=grip, Contribution, Growth, Uncertainty, Significance),
budget-writer register, repo + academic evidence, pushed immediately per
explicit instruction. Also flagged: the PMU lens misrouted this very request
to the pmu-walk/Rust-chip domain instead of a blog/content domain — a real
reliability gap in the routing lexicon, now logged in the post itself.
Persona-Intent: A budget-conscious, moderately technical reader who runs AI
agents themselves reads this cold and should walk away with: (1) a working
two-axis mental model (capability vs. grip) they didn't have before; (2) more
trust in the company, not less, because it published its own live routing
failure instead of hiding it; (3) the vocabulary (deterministic vs. decidable)
to name what they've been vaguely calling "hallucination." qwen2.5:7b cold-read
confirmed all three; flagged one confusing phrase ("memory with no physical
anchor"), rewritten before commit.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Xm3fjuwHFb9ZxVSYJDVrAL

commit 6a7f5e120 — 144×144 lattice triptych (on-chip)

◎ the ingest chose A1,C3 — Strategy·Law × Operations·Flow — acting on A3,B2 — Strategy·Fund × Tactics·Deal (grip 0.828) · the walk starts at the ACTOR's row · 5 file(s)

⏱ ingest 219.3ms (corpus→lattice) · definer walk 160ms (row→transpose→row, ballistic XOR per hop) · total 472ms
ingest here = commit-scoped SENSING only (this commit's message + changed files → lit anchors) — deep seed authoring lives in the reef-self-loop, off the commit path

rows = ACTOR (lens) · cols = PATIENT (object) · diagonal = self-reference; ◎ at actor-row × patient-col — every panel, same orientation. Heat above the diagonal = actors ranking ShortLex-earlier than their patients (the walk follows definers uphill in rank).
The 12 canonical lanes (rows top→bottom, cols left→right): A·Strategy · B·Tactics · C·Operations · A1·Law · A2·Goal · A3·Fund · B1·Speed · B2·Deal · B3·Signal · C1·Grid · C2·Loop · C3·Flow

PAIR LATTICE (the map we normally see)

PRE-WALK SENSE · INTENT grid — the snapshot, no point of view

the no-point-of-view snapshot — what the sensor lit before any walk; the leaf walk below is what bridges the two grids

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PRE-WALK SENSE · REALITY grid — the snapshot, no point of view

the no-point-of-view snapshot, reality side — same sensor law

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PRE-WALK Δ · straight comparison of the two ingested grids (green both · cyan intent-only · amber reality-only) · 38% overlap · ◎ = where the competence pixel sits in the raw overlap

good = green overlap without being a copy — near-100% is self-similarity, not alignment

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

◎ the pixel this commit belongs to

A1,C3 — Strategy·Law × Operations·Flow (actor) acting on A3,B2 — Strategy·Fund × Tactics·Deal (patient) · grip 0.828

actor seed A1,C3 — Compliance mandates establish the binding rule that regulates flow throughput and pipeline delivery rates. EU AI Act Article 14 requires human-on-the-loop oversight, with the supervisor able to interrupt the system at any moment. Compliance attaches to the artifact the artifact cannot escape — provenance is a regulatory primitive, not a feature. Constraints precede objectives. The cap-table, the license text, the consent envelope — each draws the inviolate line the action must respect, and the line is enforced before the optimization runs. A rule names what the system MAY NOT do. If the agent can do X, then 'forbid X' is not the constraint — it is wishful thinking dressed as policy. Constraints live in the gate, not in the prompt. Regulation is not advisory. The fine is the price tag stapled to the omitted control; insurance is the option pricing on the failure-mode the regulation already named. Flow is the rate at which committed work crosses the finish line. Inventory accumulates upstream; flow is the only thing the customer pays for, and the only thing the operator can measure honestly. Movement is the irreversible delivery of value across the boundary. Anything stationary is cost; anything moving is either revenue, deferred revenue, or the exhaust of revenue not yet realized. Throughput dominates utilization. A ninety-nine-percent utilized system at low throughput pays the same wages as a fifty-percent utilized system at twice the throughput, but the second one ships the customer's order. Flow has a direction. Bidirectional flow is two flows, each measured separately; calling them a single flow hides the chokepoint and the chokepoint is where the leverage lives.

patient seed A3,B2 — Sufficient capital funds provide the dollar floor needed to secure a favorable deal rate and exchange terms. Every dollar builds either the floor (substrate that lifts future bets) or the ceiling (cap on the next round). Allocation is which one this dollar chooses, and the choice is irreversible by the time the wire clears. The deal sets the exchange rate between two value substrates. Mis-priced exchange is a slow-bleed; correctly priced exchange compounds both sides, because each cycle of trade enriches the bench on both ends. Burn rate divides into substrate-building and exhaust. Substrate compounds; exhaust evaporates by quarter-end. The investor who pays for substrate is buying a different asset than the investor who pays for exhaust. Terms are the API between two organizations — every clause is a contract method that survives the handshake. Renegotiation is the breaking change; the deal is the stable interface either side can build against. Capital allocation is the irreversible vote — the bet that closes options to keep the others open. Each dollar deployed buys both the asset and the foreclosure of every alternative the dollar could have funded. Negotiation closes at the moment both parties' BATNAs cross. Push past that point and you've bought regret at retail; stop short and you've left margin on the table that the next deal will not return to you. The funding question is the runway question. Eighteen months at one burn rate is a different company than twelve months at a higher burn rate, even when the bank balance matches. A signed counterparty is more valuable than an unsigned promise. The signature is the substrate; the conversation that preceded it is the exhaust the signature converted into substrate.

INTENT · leaf-walk clouds · BLUE→VIOLET by ply (◎ = the competence pixel: actor∩patient)

good = the clouds concentrate where the commit SAYS it works

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

REALITY · leaf-walk clouds · ORANGE→RED by ply (◎ = the same pixel — the shared perspective)

good = the same clouds as intent — shipped where declared

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

DELTA · graded cloud divergence (green agree · magenta declared-not-done · amber done-not-declared) · 30% agree

good = mostly green; magenta = said-not-done (chase it in the code), amber = done-not-said (chase it in the docs)

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

TOLERANCE · cloud topology (dim green in-shape · hot amber bleed · red orthogonal concentration)

red only matters above the flip — a few amber is normal bleed

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

SHORTLEX-3 PROJECTION (ABC · A1..C3 · the 132 children — candidate, pre-ratchet)

PROJECTED INTENT · the commit’s claims, head→axis_i tail→axis_j (cyan)

the three-length view — good = claims land inside their own zone

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PROJECTED REALITY · same projection law (amber)

same projection law — good = reality occupies the zones intent does

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

STRAIGHT COMPARISON · green both · cyan intent-only · amber reality-only · 21% overlap

good = the zones agree; cross-zone scatter is the drift to read

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

shortlex-3 zone occupancy (lit cells per diagonal square)
INTENT: 2 in the 3×3 ABC corner · 5 in the 9×9 A1..C3 square · 1291 in the 132×132 children square · 269 cross-zone
REALITY: 3 in the 3×3 ABC corner · 12 in the 9×9 A1..C3 square · 1546 in the 132×132 children square · 370 cross-zone
axes: identical 144 ShortLex names both sides (symmetric); weights asymmetric
zone 3 = candidate children, pre-ratchet (repo-derived dumps not yet past the perturbation-probe gate) · intent 89 claims @ θ 0.6406 · reality 46 claims @ θ 0.625 · projection 300ms

SHAPE-MATCH σ = 17.27 — the real intent↔reality definer walk vs random-reality (actual 0.7268 vs random 0.071). >0 = the walk distinguishes real alignment from noise; aggregate over independent walks → the divergent series.
σ_drift · verified-reef · verified-reef territory — the instrument at full grip

Most out-of-lane row: C3 · Flow — that lane is drifting most (read the red band in that row).

Three-colour tolerance (out-of-lane = reality fires where intent is weak, in orthogonal blocks): green=in-lane agreement · amber=a few out-of-lane (tolerated) · red=armed — orthogonal out-of-lane 5% vs tolerance 25%.
762 green · 218 amber · 0 red · 137 XOR-friction cells.

Both axes = the 144 ShortLex anchors, cell=row⊕col, on the chip. INTENT=docs+rules · REALITY=code · DELTA=friction.

tile dump — what the SimHash matched on the diagonal (inspect the sensor; tap to expand)

A,A The lattice we build today is what next year's underwriter inherits.

intent 78% The debate itself: Ben Goertzel and Kai Zen Bickle, [AGI Experts: We're About To Lose Control o

reality 72% <div className="flex justify-end text-sm text-gray-500 mt-4">🐠🎣🧭 C → D 🌀</div>

B,B Tactics is the choice of which leverage to pull at which beat.

intent 75% (no distinctive match)

reality 75% (no distinctive match)

C,C Operations is the loop that runs whether anyone watches.

intent 80% (no distinctive match)

reality 78% (no distinctive match)

A1,A1 EU AI Act Article 14 requires human-on-the-loop oversight, with the supervisor able to interrupt the system at

intent 75% (no distinctive match)

reality 77% (no distinctive match)

A2,A2 The target is the cell coordinate the system MUST occupy by end-of-quarter.

intent 75% (no distinctive match)

reality 73% (no distinctive match)

A3,A3 Every dollar builds either the floor (substrate that lifts future bets) or the ceiling (cap on the next round)

intent 77% <div className="font-mono text-xs uppercase tracking-widest text-gray-400">the significance of

reality 73% (no distinctive match)

B1,B1 Latency is the price of a missed beat.

intent 80% The third gap is new, and we didn't have to go looking for it. The transcript itself loses grip

reality 77% <span className="text-xl font-semibold">Ahead of the timeline debate: arguing the right axis</s

B2,B2 The deal sets the exchange rate between two value substrates.

intent 78% (no distinctive match)

reality 80% (no distinctive match)

B3,B3 Signal-to-noise is the inverse cube of bandwidth — narrow the channel, the message arrives; broaden it, recipi

intent 70% (no distinctive match)

reality 70% (no distinctive match)

C1,C1 Power flows where the grid lets it.

intent 78% <div className="font-mono text-xs uppercase tracking-widest text-gray-400">the certainty worth

reality 78% (no distinctive match)

C2,C2 Each iteration is a hypothesis tested against the ground truth of the world.

intent 78% (no distinctive match)

reality 80% (no distinctive match)

C3,C3 Flow is the rate at which committed work crosses the finish line.

intent 75% (no distinctive match)

reality 75% import BlogCallout from '@/components/BlogCallout';

commit context — 5 file(s)

blog: Generally Correct, Specifically Wrong — the grip problem AGI timelines skip

Reframes hallucination as a domain-placement failure, not a data error: general
capability and domain grip are different axes, so recursive self-improvement
without a grip instrument compounds drift instead of curing it. Responds to a
recent long-form AGI-timeline debate (hedged honestly — transcript was not
independently retrievable) betting on next-token-predictor + external-memory
architectures closing the gap to AGI.

Includes a real, unplanned example: our own prompt-time lens misrouted THIS
exact request (write a blog post) into the on-chip Rust walk domain instead of
a content domain, live, mid-composition — the cleanest possible evidence for
the post's own thesis, logged as a routing-reliability gap rather than edited
out.

Originating-Terminal: 🎤 voice
Relevant-Rooms: 🎩 operator, 🧪 laboratory, 🔨 builder
Story: Operator watched a long-form AGI debate and wanted the repo's
decidability/domain-grip thesis applied to it directly — hallucination
reframed via "generally correct, specifically wrong," six needs mapped
explicitly (Connection=grip, Contribution, Growth, Uncertainty, Significance),

docs/comms/self/approved/2026-07-03-gtm-close-requirements-and-weekend-gets.grade.json
docs/comms/self/approved/2026-07-03-gtm-close-requirements-and-weekend-gets.md
docs/comms/self/sent/2026-07-03-gtm-close-requirements-and-weekend-gets.grade.json
docs/comms/self/sent/2026-07-03-gtm-close-requirements-and-weekend-gets.md
src/content/blog/2026-07-03-generally-correct-specifically-wrong.mdx

Produced on the chip by the running pipeline (scripts/pmu/pipeline.mjs → runPipeline → walk + xor), commit-scoped · on-chip per-walk 388.70ns (this commit). Recompute: node scripts/pmu/commit-triptych.mjs --commit 6a7f5e120

Commit artifacts (open on a computer — Gmail mobile won't follow file:// links; the inline images above are the phone-viewable form): reef grid · this triptych · the HDL — all three are also attached to this email.