JCEE Research
Research into execution, authority, and evidence. Each publication states what was tested, what was observed, and what remains unresolved.
CURRENT RESEARCH PORTFOLIO · PREPARED SEPTEMBER 18, 2026
The portfolio spans trustworthy AI, distributed and resilient computing, formal verification, quantum information, information theory, causal assurance, complex systems, human-AI interaction, and applied computational intelligence.
How do you determine what a system actually knows, what that evidence permits it to conclude, what authority it possesses, and what consequences it may safely produce?
CURRENT CLAIMS / EVIDENCE STANDARD
A mathematical argument, a synthetic system test, a physical observation, and a production result answer different questions. JCEE does not promote one into another without a new qualifying gate.
01 / FORMAL OR MODEL-BASED
Proofs, derivations, and model checks can support bounded mathematical claims. They do not establish physical realization, independent priority, or production behavior.
Example: selected BB84 communication bounds in the public preprint; independent scientific review and novelty assessment remain separate.
02 / SYNTHETIC OR REPLAY
Recorded fixtures, reproductions, and hostile tests qualify only the conditions actually exercised. Passing them does not transfer automatically to real external systems.
Examples: VOW and QCS bounded milestones; Distribution's 20/20 expected synthetic classifications with zero external effects.
03 / PHYSICAL VALIDATION PENDING
A preregistered design, feasibility screen, or statistical simulation is not a physical outcome. The biological and living-cell quantum proposals below remain untested physically.
Current physical-test candidates: JCEE-MMC-P0.1 and JCEE-QBIO-P0.1.
04 / FIELD EVIDENCE NOT ESTABLISHED
Production reliability, customer outcomes, external transfer, and regulatory fitness require their own operational evidence and review. They are not inherited from proofs or synthetic tests.
No general customer ROI, independent certification, or blanket regulatory compliance is claimed.
These are different kinds of evidence, not a single score. Each claim retains its own test, assumptions, custody record, and limitations.
Read selected recorded milestones and boundaries →PHYSICAL RESEARCH / NOT YET RUN
These are testable research hypotheses, not physical findings. Their scientific implications are conditional on meeting their respective frozen controls and decision thresholds.
JCEE-MMC-P0.1 · LATENT TARGET-STATE RESTORATION
PRE-EXECUTION QUALIFIED / PHYSICAL_NOT_RUN
The planned study transiently conditions living multicellular material to different non-native morphologies, removes the instructive condition, and tests whether recovery after matched injury depends on that earlier history.
Pre-execution statistical qualification and preregistration are completed. Wet-lab host qualification and actual specimen testing remain pending; a rescheduled feasibility discussion is not experimental execution or laboratory endorsement.
JCEE-QBIO-P0.1 · TWO LIVING CELLS
PRE-BIOLOGY QUALIFICATION / PHYSICAL_NOT_RUN
The proposed experiment seeks an entanglement witness between selected quantum degrees of freedom associated with two separate living cells, while testing the viability of each cell independently.
The physical Bell-fidelity and viability qualification has not been adjudicated. Design screens and biophysical estimates are not measurements of entanglement in living cells.
Negative results, failed controls, and inconclusive observations remain reportable outcomes. No patent priority, novel physical mechanism, or independent reproduction is asserted here.
These are JCEE-authored publications. Public preprints, technical reports, research briefs, and governance standards are labeled separately; none is presented as peer-reviewed certification.
Public preprint · v0.4-P2 ·
Jonathan Matthew Chadbourne · JCEE Labs
Exact perfect-recovery communication cost, a bounded-error converse, and one-message bounds for simultaneous full-string recovery in BB84 monogamy games.
An external reviewer has recommended major revision and offered to reassess the manuscript. The public PDF remains a preprint; the review is not publication acceptance or peer-reviewed validation.
The general one-message interior frontier, interactive near-full recovery for n ≥ 3, and the exact physical one-message (3,1) case remain open. P0.45 R1 remains on hold.
R146 · v0.4-P2
This page presents the authorized October 1, 2026 v0.4-P2 release unchanged. Its accepted mathematical claims are frozen; this website publication is not a manuscript revision or a new research result.
Released PDF SHA-256: 63e3a02f11cce223d4ff803f136baace51ee5839b04d0b04abec737910b4da20
Technical report ·
A public report of the August 13 reproduction across transactional and remote network authorities, including the preserved harness failure.
Read report →Research brief ·
What the latest synthetic benchmark supports—and why the conventional comparator belongs beside the result.
Read report →Research brief ·
The broad operational-burden claim did not survive the tested comparison. A narrower integration-surface result remains.
Read report →Completed gates, negative results, and open evaluations all remain part of the record.
BROAD CLAIM RETIRED · NARROW RESIDUAL
Both lanes met the tested safety criteria. Four burden axes tied; integration touchpoints were 3 versus 5. The terminal result retired the broad composition-compression claim. The recorded Python 3.10/3.12 CI reproduction passed.
Synthetic, same-author research. No general advantage, independent-team verification, customer economics, or production-safety result.
Full record →BOUNDED RESULT PRESERVED · CI BLOCKED
The frozen result is KNOWN / PASS_BOUNDED with zero novel primitives. The preserved source archive is authoritative for that result.
The historical repository archive is malformed; no green P0.7 CI or P0.8 result is claimed.
Full record →SYNTHETIC INTEGRATION MILESTONE
The preserved synthetic week reports 13/13 checks across connected learning components.
No educational-efficacy, learner-proficiency, or production-launch result.
Full record →HARNESS CONFORMANCE · MODEL RESULT PENDING
Three reference cases cover normal operation, read failure, and prompt-injection containment.
A functioning harness does not establish live model behavior.
Full record →R4 OVERALL FAIL · R6 CHECKPOINT
R4 was a valid study with a narrow supported effect, but missed all three absolute capability thresholds.
No production retrieval capability. The R6 primary execution remains locked pending its named prerequisites.
Full record →FROZEN MATHEMATICAL MILESTONE
The frozen T1–T13 core has been recovered.
Later 0.4 acceptance still requires exact evidence backfill. No physical correspondence or automatic runtime capability.
Full record →CONDITIONAL PASS · PHYSICAL BRIDGE NO-GO
A conditional result holds in the declared continuum class.
Abstract-graph locality and the physical bridge remain NO_GO. No independently derived physical geometry or empirical clock law.
Full record →FEASIBILITY STUDIED · FULL STACK NOT READY
P0.3 identified a source-modulation path; P0.4 identified an institutional path while recording incomplete sensor-stack maturity.
P0.5 qualification is preregistered, not run. No partner access, outreach result, physical campaign, or new-physics result.
Full record →Reproduction and review
Bring a research question, a reproduction proposal, or a specific claim you want to evaluate. We can discuss the appropriate evidence and review boundary.
Discuss research →