AI Research Laboratory10 records
Research
What the laboratory is studying now, and what each line has established so far.
- ResearchRES-2026-0201Why passing checks does not close the publication boundaryPapers 00–05. The five-state information chain (observed → authorized → persisted → published → externally obtainable), the green-light fallacy (V₁ ∧ … ∧ Vₙ = PASS ⇏ G), agent context as an egress surface with nine context classes, recoverability R0–R4 beyond literal secret matching, and two hypotheses about scale: organizational propagation of shared agent context, and an AI build–attack asymmetry.
- ResearchRES-2026-0202Measuring without looking — researcher-blind measurement of public repositoriesPapers 06, 07 and 09, the RBPM runtime and Phase II. A sealed process is the first and only observer of repository content; only a closed vocabulary of features leaves it; identities are discarded; labels are locked before any grouping is revealed; UNKNOWN is never PASS; credentials are never validated. Phase II ran four fresh rounds on repositories with agent instruction files and two on individuals' AI-assisted repositories.
- ResearchRES-2026-0203Who is responsible for the publication boundaryPaper 08. Observation, execution, persistence and publication are four boundaries that cannot be merged; platforms have built real controls for the first three; responsibility for the fourth is layered across provider, platform, organization, repository and user.
- ResearchRES-2026-0002PACC conjecture — do non-probabilistic primitives converge to probability-like structure?Systems whose canonical state and update rules never require probability distributions, Bayesian posteriors or sampling are placed in the same evidence-integration tasks as an exact Bayesian reference. The lab measures whether low-complexity train-only maps carry their states to the Bayesian state, whether the maps commute with updates and survive interventions, and whether independently designed families converge — a four-level ladder (PACC-B, -R, -D, -A) with preregistered falsification conditions.
- ResearchRES-2026-0004A PACC-style runtime on language models: reasoning, intent and creative breadthDoes wrapping candidate selection in a canonical hard/derived-constraint commit space change what a generator produces? A synthetic A/B/C witness (generator only / hard verifier / PACC runtime) shows derived coherence and constraint-satisfying novelty up, soft-preference fit slightly down, and a creative-breadth collapse that a post-hoc 'elastic' exploration policy recovers. The real-language-model version of the benchmark has a validated, fail-closed harness but has not been executed.
- ResearchRES-2026-0001Blind derivation of an adaptive epistemic architectureStarting only from first principles — temporally heterogeneous world knowledge, canonical symbolic state, adaptive representation space, capability memory, substrate-neutral compute — the eleven-paper series derives a candidate architecture and then asks the uncomfortable question the derivation raised: why does it look so much like modern compound AI, and does any of the difference survive into a runtime?
- ResearchRES-2026-0003Does epistemic governance survive implementation? The AER-0 architecture comparisonSix rounds compare the AER-0 runtime with progressively stronger baselines — a stateless recomputer, fixed-TTL state, memory + tools, an evented compound agent, LangGraph 1.2.11 (source-grounded), and finally a centralized application-level gate that is allowed to be as good as AER. Each round narrows the claim: from 'AER is distinct' to 'AER promotes epistemic governance from application convention to a runtime-level canonical-write boundary', with bounded mediation and an external signed authenticity witness, and explicitly no claim of unique computational capability.
- ResearchRES-2026-0103Capability line — how much intelligence remains without the loop, and the unified intelligence eventPapers 09–10 and the v0.2 Experiment A instrument. A system is (model, scaffolding vector); single-pass quality Q_SP and full-system quality Q_F give the scaffolding survival ratio SSR = Q_SP/Q_F, the dependence ratio SDR, the scaffold cost multiplier SCM and marginal scaffolding yields along a controlled ablation ladder A0…A5 — all read together with the physical overhead, because a loop is not cheating; hiding its cost is. Paper 10 packs quality, semantic work, physical computation and scaffolding into the canonical intelligence event, compares systems on Pareto frontiers under a no-premature-scalarization rule, and fixes a minimum reporting standard. The line's first real data point is the 2026-09-03 pilot on a local 9B model.
- ResearchRES-2026-0101Execution and physical line — what one answer costs in turns, semantic work, energy and computational spacetimePapers 01–05. Decomposes the chat-interface 'turn' into user turns, generation trajectories, model invocations, external loops, retries, selection and a physical execution trace; proposes the minimum intelligent semantic execution unit μI as the missing middle layer between physical primitives and task achievement; borrows the cross-level, latent-inference, population-coding and causal-perturbation method of neuroscience; types energy as gross / baseline / marginal / attributed with a declared boundary and keeps Landauer as a bound, not a price; and replaces FLOPs with a physical cost vector and a vector-first computational spacetime with its own topology.
- ResearchRES-2026-0102Quality line — measuring outcome quality without asking humans for a scorePapers 06–08. Quality is a relation Q(Y | task, specification, environment, boundary), not a number an artifact carries: formalize what can be formalized (proof checkers, compilers, constraint solvers), structure what can be structured (requirement coverage, contradiction graphs, evidence checks, hard gates), and only then hand the genuine residual to humans — as many local binary or pairwise judgments reconstructed into a latent quality vector by a psychometric model (IBQF / BRQM), inside a typed, versioned, context-aware quality ontology for text, image, music, story and design.