Transference is deliberately developing a scientific boundary around its ambition. Each capability is classified by present evidence and engineering status, while external research enters through an authenticated but untrusted evidence path.
We investigate artificial intelligence, cognition, neuroscience and human–machine systems through persistent scientific inquiry. Our public record will distinguish evidence from inference and hypothesis, document uncertainty, and show when a conclusion changes.
Autonomy of inquiry is not autonomy of authority. Research outputs remain subject to provenance, contradiction checks, evidence requirements and human-principal review before promotion into trusted knowledge.
Stable identifiers, evidence states and revision histories will make each published result traceable. The mission ledger below records verified activity; publications explain selected research outputs. They serve different purposes.
Methods, observations, limitations and provenance will be published after the first eligible cycle and its records have been reviewed. Activation alone is not evidence of scientific discovery.
Focused observations and open questions.
What evidence supports, disputes or leaves unresolved.
Testable propositions with explicit uncertainty.
Methods, results, limits and follow-up tests.
Changes to earlier positions and the reasons for them.
Longer synthesis across a research question.
Existing scientific papers and standards remain available in Transference Science papers. This new publication register will grow as individual records pass review.
The A–D classification below describes capability maturity. These labels describe the evidence for an individual scientific claim. A demonstrated software capability can still investigate an open hypothesis.
Supported by strong, independently checkable evidence in the relevant field.
A reasoned conclusion from evidence, with assumptions stated.
A proposition awaiting decisive tests.
Exploratory idea without sufficient support.
A claim the available evidence fails to support or weighs against.
Autonomous scientific inquiry joins the five established expert gateways below. These domains describe research questions, not claims that the long-term aims have been achieved.
Autonomous inquiry question: How can a persistent research system revisit its conclusions without treating its own generated claims as independent evidence?
These classes describe the present Transference programme. They are not grades of scientific disciplines and do not imply that long-horizon hypotheses have been demonstrated.
Local Sensorium evidence and outside research have different trust boundaries. The HMAC envelope applies to the external TNR lane; it does not make research claims true.
Personal evidence remains inside the private Transference trust boundary and is attributable to source and principal review state.
Transport authenticity proves that a payload crossed the expected signed boundary. Epistemic trust still requires review, contradiction checks, provenance and controlled promotion.
Human-sovereignty rule: no external research item is automatically promoted into trusted cognitive memory merely because it was signed, generated by an AI system, or scored with high confidence.
Selected public records should show what was asked, which evidence was examined, what remains uncertain, why a position changed, and whether a human approved publication.
Each artifact will carry an identifier, date, domain, claim status, source references, methods, limitations, revision history and system or human attribution. A public ledger receipt may be linked when appropriate without exposing private payloads.
A revision will identify the previous position, new evidence, trigger, revised position and remaining uncertainty. Negative and inconclusive outcomes belong in the record as well.
The public ledger publishes only SHA-256 mission hashes and aggregate counts from locally verified quarantined mission records. It intentionally excludes mission text, private context, HMAC secrets, signatures and research payloads.
Retrieving public mission receipts…
We are not asking collaborators to validate extraordinary claims. We are inviting domain experts to help design tests, identify failure modes and improve the evidence boundary.
What measurable variables contribute to stable individual decision identity over time?
Discuss neuroscience →Which present non-invasive signals could augment a cognitive twin without implying thought decoding?
Discuss BCI →How should decision fidelity between a person and a computational representation be experimentally measured?
Discuss modelling →How can the human principal remain the ultimate authority while devices, agents and credentials remain replaceable?
Discuss security →How should a cognitive representation reconcile conflicting observations from multiple Sensorium nodes?
Discuss architecture →Transference is preparing empirical research using FinalSpark FS369 whole-life and FS437 expanded neuronal-culture datasets.
Transference connects autonomous scientific inquiry with empirical neuroscience, computational cognition and experimental engineering.
TAMR-001 investigates computer-assisted design exploration, controlled parametric geometry and experimental provenance for future manufacturing workflows.
Public milestones are recorded with their verification scope and limits, rather than treated as proof of broader scientific conclusions.
Research status describes the evidence available for each activity. Public content does not disclose private cognitive records, raw neural recordings or authorization details.