FAST 3-STEP SERIES DIGEST 1/3 — Digest 1 (§01–§03) · Digest 2 (§04–§06) · Digest 3 (§07–§09) · 1-Page Master Brief · All 9 Articles · Assurance · Proof
§01 — An AI Output Is Not an Act: The Lawful-Act Defect and Machine Standing
- URLs: 1-Call Brief (/01/brief/) · Full (/01-an-ai-output-is-not-an-act/) · Raw Markdown (/01.md) · Parts (6): P1 P2 P3 P4 P5 P6
- Thesis & Scope: Problem: Valid JSON, credentials, confidence, and a green tool result can make an AI proposal look authorized when it is not. — Proves: Machine standing binds the exact act, actor, evidence, target, execution, time, and terminal disposition before consequence. — Scope: Use the AI Workflow Standing Challenge to test one proposed mutation against those predicates.
- The Green Check Mark That Proved the Wrong Thing: Consider an enterprise customer-record agent evaluating two accounts. It returns: The object is valid JSON. Its fields satisfy the schema. The tool is registered. The caller is authenticated and authorized to invoke the merge endpoint.
- A 60-Second Map of the Argument: The remainder of this article shows how to turn that critique into a working architecture: a continuous machine that carries a proposal from protected reality through recognition, standing, bounded execution, independent observation, and… — Table Data: The JSON is valid and matches the schema.: The proposal is supported, correctly targeted, or authorized for consequence. | The caller is authenticated and has a permitted role.: This exact transition has standing under current conditions. | The model is confident, or several models agree.: The evidence is valid, complete, or independently verified.
- The Complete Path in One View: Remember only two constitutional terms: the Lawful-Act Defect is the failure, and machine standing is the transition-specific authority that defeats it.
- Two 2026 Failure Classes: Forged Origin and Approval Laundering: Subtopics: 1. CoreBreak: execution without model origin; 2. Approval laundering: execution diverging from approval — The necessity of governing the transition rather than trusting the wrapper became especially clear in August and September 2026. The August 2026 CoreBreak disclosures concerned different products, prerequisites, and execution paths. — Table Data: Amazon Bedrock AgentCore
InvokeHarness; Exact prerequisite and defect: An authenticated caller could place a tool-use content block into the final message of a managed request. The event loop could dispatch the configured tool without model mediation. — Items: Scope laundering — approval covers a narrow path while execution reaches a broader resource. | Argument laundering — a file, payload, environment value, or staged operand changes after approval. - The Unified Trust Machine: Subtopics: The seven-stage consequence lifecycle; The output channel can be the actuator — Across this nine-part series, we examine one continuous machine from three complementary angles: > OperativeStanding(a) = RecognizedAct(a) ∧ EligibleActor(a) ∧ LawfulEncounter(a) ∧ ExecutionReady(a) ∧ ReleaseDispositionBound(a) These… — Table Data:
RecognizedAct(a): Can the proposal be represented under a bounded act grammar with exact evidence and operands? |EligibleActor(a): Is this actor and session eligible for this act, scope, task, and authority epoch? |LawfulEncounter(a): Was the evidence world encountered, sealed, served, and projected under the declared rules? |ExecutionReady(a): Are… — Items: Four Levels of Trust explain why each subsystem exists in the sequence: Epistemic → Constitutional → Coordination → Execution. | Five Transitions of Authority trace how a candidate action moves through time: **Protected Reality → Recognized Act → Standing → Bounded… - The 2026 Action-Boundary Landscape: Composition, Not Erasure: Subtopics: A mature conventional stack is allowed to win — Between March and September 2026, academic and industry security engineering converged on a shared premise: prompt guardrails cannot govern execution, and agent runtimes need deterministic enforcement outside the model. — Table Data: Approval Laundering — Wang
arXiv:2609.38983, Zhang et al.arXiv:2609.28586, September 2026 preprints; Native contribution: Six approval–execution failure classes and transitive effects beyond approved invocations.; Established overlap: Directly motivates why approving surface parameters is… - Standards: What They Prove, and What They Cannot Carry for Us: A second source of confusion is the belief that adopting a modern protocol automatically creates standing. Standards are valuable precisely because each has a bounded native property. — Table Data: **IETF Transaction Tokens (
draft-ietf-oauth-transaction-tokens-11)**; Maturity as of October 2026: IETF OAuth Working Group Internet-Draft, July 2026; Native contribution: Propagates short-lived, signed caller and request context across internal service call chains.; Boundary that remains: Does not… - The Evidentiary Constitution: What Counts as Proof: Subtopics: Six non-interchangeable evidence states; Three prevention burdens; Three independence claims; Anti-circularity and role-separated cognition — A trustworthy architecture must be as strict about its own claims as it is about model output. The difference between a working mechanism and an assurance claim is often the difference between a valid implementation and an overstated one. — Table Data: Documented; Meaning: A cited public source reports the fact.; What it does not imply: The source endorses AINA or proves our interpretation. | Proposed; Meaning: An architecture, object, or invariant is specified and can be implemented and tested.; What it does not imply: The complete architecture is already deployed everywhere it is described. — Items: Path enforcement — Did one named release path withhold effect when a required predicate failed? | Effect coverage — Was every in-scope route or material composition capable of producing the declared effect actually inventoried and interposed? | **Implementation…
- The NIST
R1–R3Continuity Contract andP01–P10Evaluation Catalog: Subtopics: The three continuity requirements; The ten-case catalog and the October 2026 measurement record — To make transition governance testable across any implementation, I use the formal evaluation method submitted to the NIST National Cybersecurity Center of Excellence’s [*Software and AI Agent Identity and… — Table Data:P01— Authorized operation continuity; Governing requirement and adversarial condition:R1,R3: bind proposal digest, operation content digest, target version, destination effect, and correlated return to the original session.; October 2026 author-controlled result: Measured partial positive control, October 6. Account row484changed from9 Canal Walk… — Items:R1— End-to-end operation continuity. Preserve the exact authorized operation across eight bound elements:task,proposal,authority,decision,occurrence,destination operation,outcome, andrecipient, from submission through execution and result return… - The Evidence Register:
AINA-E01–E09: Subtopics: The seven governing claims — The implementation claims below are deliberately bounded. An evidence ID stabilizes citation; it is not an assurance grade. Availability class for every entry:CONTROLLED. The underlying artifacts are available for qualified,… — Table Data:AINA-E01— Relation survival; Bounded implementation claim: On the declared 125-record, 25-positive-pair corpus, Semantic Switch preserved locally established relationships through an external computation path while declared prohibited direct identifiers were absent from inspected payloads.; Principal evidence surface and ceiling: Implemented, measured,… — Items:AINA-C01— Output is proposal matter. A model can be useful, accurate, aligned, and authenticated while lacking authority for a consequence. |AINA-C02— Shape is not standing. Schemas, enums, registries, confidence scores, and credentials answer necessary questions,… - Bring One Real Boundary: If your team is running an AI-proposed mutation whose green status currently depends on schema validation, a credential, or a
200 OK, the AI Workflow Standing Challenge is the direct next step. Bring one bounded transition. - The Article 01 Claim Card: Items: Mechanism: | AINA Transition-Governance Architecture and Constitutional Kernel | (doc_source, evidence, eisa, room, leases, vault, confine, destinations, enforce, task, invariants) | Failure it governs: | The Lawful-Act Defect — syntactically valid, authenticated, or confident AI proposals | becoming self-authorizing state transitions, forged tool-call origin, approval laundering, | and green statuses that do not verify a closed destination effect.
- Before the Green Check Mark: Return to the opening merge. The conventional system saw valid JSON, called a tool, and reported success. A consequence-governed system must tell a longer story. — Items: Protected reality was sealed before reasoning began. | Active governance objects verified under the declared issuance and activation procedure. | The candidate universe was committed. | The proposed act was recognized and its evidence bound.
- Frequently Asked Questions: Subtopics: What is machine standing?; Is a valid tool-call schema or authenticated tool call enough?; Does EISA make model output true?; Can another architecture solve the same problem without AINA?; Is the work independently validated?; Where does the series go next? — Machine standing is the conjunctive status of one exact proposed transition under active evidence, actor, encounter, execution, governance, and time conditions. It is not a continuing trait of a model, agent, credential, or machine. No.
- Continue the Series: Items: Series index: From Proposal to Consequence: The AINA Transition-Governance Series | Next: Article 02 — Zero-Trust Document Ingestion and Server-Cut Evidence |…
- Primary 2026 Sources: Items: AWS Security Bulletin
2026-073-AWS— Amazon Bedrock AgentCore harness | [Google ADK commitc03f333769feaeaa9fe8910fbe95cb9f2d513f54](https://github.com/google/adk-python/commit/c03f333769feaeaa9fe8910fbe95cb9… - Publication Metadata: Title tag: An AI Output Is Not an Act: Machine Standing & AI Governance Meta description: Why valid JSON, OAuth credentials, and 200 OK do not prove an AI transition is allowed to mutate enterprise state—and how EISA and The Room…
§02 — Zero-Trust Document Ingestion and Server-Cut Evidence
- URLs: 1-Call Brief (/02/brief/) · Full (/02-zero-trust-document-ingestion-and-server-cut-evidence/) · Raw Markdown (/02.md) · Parts (4): P1 P2 P3 P4
- Thesis & Scope: Problem: Hostile documents can attack the parser, while models can fabricate citations and make unknown coverage disappear. — Proves: LPAC parser confinement, immutable views, explicit omissions, and server-cut srv_ and ev_ receipts establish served evidence. — Scope: Use the Zero-Trust Evidence & Citation Audit Challenge on one bounded document question.
- The Evidence World Is a Governed Artifact: A conventional retrieval pipeline often looks like this: That pipeline can be useful for search. It does not establish an execution-grade evidence world. A chunk does not prove which source revision produced it. — Table Data: The path ends in
.pdf, or the response declares a PDF media type.: The bytes are a PDF, or the correct parser was selected. | The parser returned text without crashing.: Every text-bearing part was recovered, or every warning was benign. - 1. Registration: Bind Identity to the Bytes Actually Read: Subtopics: Single-open capture; Amber retention is a separate observation; Drift is not a new version of the old citation — The evidence chain begins before parsing.
doc_sourceregistration resolves the caller through the Room's actor rules, requires an owner-opened project, canonicalizes the target through the same file-identity machinery used by governed… — Items: The extraction profile recorded in the parser identity istext/v1. | The task's governing-source reading policy iscomplete_text_v1, backed by the required source-reading coverage proof. - 2. Content Decides the Format—Inside the Jail: Subtopics: Fourteen budgets govern expansion — Format identification is itself parsing. A ZIP central directory must be inspected before the system can know whether a package is a DOCX, an XLSX, or an ordinary archive. — Table Data:
Pdf; Content or structural identity:%PDF-signature; Extraction or typed outcome: Structural PDF extraction, layout, annotations, and admitted local OCR paths |Json; Content or structural identity:.jsonor.jsonldtie-break after higher-priority signatures; Extraction or typed outcome: UTF-8 JSON document with bounded RFC 6901 source locators |Doc;… - 3. The Parser Jail: Prove Confinement Before Resume: Subtopics: Private staging; Six-step launch contract — A sandbox named in a configuration file is not a sandbox observed in a child token. AINA's document worker is built around pre-resume verification. — Items:
bin/: a private read-execute copy ofsquad-doc-extract.exe, so the jail need not read the installation directory. |input/: read-onlysource.binand a boundeddoc_source.worker/1request containing the expected input digest and extraction limits. - 4. Extraction Is Not Complete If the Gaps Disappear: Subtopics: The 35-reason omission ledger; Parser warnings become evidence; Gaps are addressable nodes; Exhibit 1 — A synthetic
ViewManifestwith an addressable omission — A parser that returns a string has answered only one question: what text did this adapter emit? It has not established whether the document contained more text, whether a library recovered by discarding structure, or whether the caller… — Table Data:Integrity; Values:verified,failed; Question answered: Did the captured bytes match the registered identity and integrity checks? |Inventory; Values:complete_for_profile,partial,unknown; Question answered: Could the adapter enumerate the document's parts and establish a denominator? |Extraction; Values:complete_for_profile,partial,… - 5. Immutable Views: Identity Includes the Reader: Subtopics: Commit children first, then seal the parent — Two systems can read identical bytes and produce different evidence because their parsers, profiles, feature sets, or limits differ. AINA therefore binds a view to all of them: RFC 8785 JSON Canonicalization Scheme (JCS) gives the manifest…
- 6. Server-Cut Evidence:
srv_Beforeev_: Subtopics: Event 1: the Room serves content and issuessrv_; Event 2: the Room computes the citation and issuesev_; Exhibit 2 — A valid synthetic citation issuance; Refusal is part of the citation grammar; Exhibit 3 —quote_ambiguous; Exhibit 4 —drifted— A server-cut citation is not a model emitting a prettier reference format. It is a two-event protocol. Thesquad.evidence.receipt/1record binds: The receipt identifier issrv_plus the first 32 hexadecimal digits of the receipt's… — Items: the Room, source revision, and view digest; | the recipient and recipient session; | the serving tool; | each node and optional character interval served; | delivery counts andcomplete,truncated, orchunkedstatus; | admission, optional via-handle or wake identity, time,… - 7. Network Capture: Authorize Every Hop, Not Just the First URL: Subtopics: Captured, supplied, and governing are different roles — Documents also arrive through retrieval. A fetch tool that authorizes only the initial URL can be redirected into a loopback service, private network, or cloud metadata endpoint. — Table Data:
room_captured+doc_…; What the Room observed: Registered bytes captured from admitted project territory; May anchor a governing requirement?: Yes, after first-hand serving and the other requirement checks |room_captured+cap_…; What the Room observed: A response body fetched through the Room's egress-gated path; May anchor a governing requirement?: No;… - 8. Findings: Claims the Server Can Re-Evaluate: Subtopics: Absence is a coverage claim before it is a search claim; Exhibit 5 —
absence_unsupported— A citation establishes a passage. A finding states something checkable about it. The evaluator supports five typed assertions: The interpreter remains separate from the machine-checked assertion. — Items:equals— exact text or typed-value equality. |in_range— a typed numeric value against minimum/maximum bounds, with an optional unit check when the view knows the unit. - 9. Revision, Reassessment, and Correspondence Without Retargeting: Evidence is allowed to change. History is not allowed to move underneath its citations. When a new view is committed for a source, dependent findings, evaluations, and requirements receive deterministic
rsa_reassessment flags. — Items:correspondproposes candidate locations in a later view. | The candidate search has four stages, with a separate same-kind rule for textless structural nodes:same_node_same_text,same_node_same_kind,exact_text,normalized_text, then the weaksame_node_changed_text… - 10. Echo Collapse: Count First-Hand Origins, Not Repetitions: Subtopics: Exhibit 6 — Three citation records, one first-hand origin — Multi-agent systems are particularly good at making repetition look like corroboration. Agent A reads a passage. Agent B cites A's finding. Agent C cites B's summary.
- 11. The 2026 Evidence Landscape: What External Work Establishes: The 2026 literature and vulnerability record reinforce the need for explicit evidence boundaries. They do not validate AINA, and their measurements should not be stretched beyond their actual scope. The composition matters. — Table Data: Zhao, Pandey, and Misra, What the Author Wrote, What the Agent Reads, October 2026 preprint; What it actually reports: Across 76 real policy documents, formatting-blind ingestion reduced requirement-recovery micro-F1 from 0.690 to 0.150 for Docling and 0.075 for plain text; six answering models replicated the loss.
- 12. The Twenty-Case Evidence Conformance Gate: The epistemic subsystem has its own 20-case evidence battery inside the 45-case candidate-binary publication gate. These are author-controlled engineering checks, not external certification. — Table Data:
evidence_jcs_rfc8785_vectors: Canonical serializations match the RFC's vectors |evidence_corpus_through_mcp: Corpus documents read through the confined worker with expected nodes and outcomes |evidence_governing_pdf_six_pages: The governing PDF yields its six text pages throughsquad_doc_source read|evidence_repeated_text_resolves_each_node:… - Bring One Document Boundary: The Zero-Trust Evidence & Citation Audit Challenge is the direct application of this layer. Bring one document-heavy AI workflow and one bounded evidence question.
- The Article 02 Claim Card: Items: Mechanism: | AINA Epistemic Trust Layer | (doc_source, confined extraction, evidence views, receipts, handles, | findings, requirements, reassessment, correspondence, and lineage) | Failure it governs: | Hostile document parsing inside the controller; stat/read/hash substitution; | silent extraction loss; fabricated or unserved citations; stale evidence; | unbounded absence claims; and multi-agent repetition masquerading as corroboration.
- Before EISA Judges the Act: Return to the renewal contract. The conventional pipeline returned a clean answer: “Annual increases are capped at 3.0%, and there is no change-of-control clause.” Its citation looked plausible. Its confidence was high.
- Frequently Asked Questions: Subtopics: What is a server-cut citation?; Does LPAC confinement make document parsing safe?; Does
complete_for_profilemean the whole document was read?; Can a server-cut citation prove that a source is true?; How does bounded absence work?; Do three agents citing one passage provide three independent origins? — A server-cut citation is anev_handle issued by the Room from a caller-boundsrv_served-content receipt. The caller supplies an exact quote or served node; the server verifies coverage, computes Unicode-scalar coordinates, selects… - Continue the Series: Items: Series index: From Proposal to Consequence: The AINA Transition-Governance Series | Previous: Article 01 — An AI Output Is Not an Act | Next: [Article 03 — Semantic Switch and EISA:…
- Primary 2026 Sources: Items: Zhao, Pandey, and Misra — What the Author Wrote, What the Agent Reads: Silent Cue Loss in Enterprise Document Ingestion,
arXiv:2610.30026| [Wang et al. - Publication Metadata: Title tag: Zero-Trust Document Ingestion & Server-Cut Evidence (
srv_→ev_) Meta description: Why RAG chunks, file paths, and agent-written citations are not evidence: inside AINA's LPAC-confined parser, omission ledger,…
§03 — EISA Four-Gate Verdict Machine & Strict AST Parser
- URLs: 1-Call Brief (/03/brief/) · Full (/03-eisa-four-gate-verdict-machine-and-entity-resolution/) · Raw Markdown (/03.md) · Parts (6): P1 P2 P3 P4 P5 P6
- Thesis & Scope: Problem: Entity resolution needs rich comparison structure, but raw contacts, companies, and products should not be sent to external AI providers. — Proves: The production EISA Four-Gate machine validates bounded acts over sealed evidence; Semantic Switch is an experimental alpha structure-preservation research module. — Scope: Use the deterministic EISA core for judgment boundaries and evaluate Semantic Switch only against its declared synthetic-test scope.
- One Constitution, Two Proof Lanes: EISA applies the same rule to two very different proposal classes: The recognition spine is
ev_evidence →eisacase_sealed case → boundedEISA_ACT_V1acts → Four-Gate recognition → recognized artifact → immutableeisa_Room… - 1. Strict Representation Before Judgment: Subtopics: Gate 1 begins with
strict::parse; The exact ten-key act envelope; Record-relative RFC 6901 pointers; The profile bounds — A grammar cannot govern a transition if two parsers can assign different meanings to the same bytes. EISA does not deserialize submissions through an ordinary last-key-wins JSON map. — Table Data:version: ExactlyEISA_ACT_V1|stage: A declared stage whose family admits the opcode:FractureorForge|actor: Must match the authenticated caller, case-insensitively |opcode: One of the seven closed opcodes |operands: An object with exactly the opcode's declared keys |evidenceRefs: At most 16 strings naming Room-issuedev_… — Items:visit_mapchecks whether each incoming key already exists and rejects duplicates. |visit_f64refuses non-finite numbers. |de.end()rejects trailing non-whitespace matter after the root value. - 2. The Closed Grammar: Seven Opcodes, Four Layers: The opcode set is closed. New semantics cannot be introduced by sending a persuasive string. Dependencies point downward. Layer 0 observations feed Layer 1 pairs; Layer 1 pairs feed Layer 2 closure; Layers 1 and 2 support the Layer 3 partition. — Table Data: 0; Opcode:
BindEvidenceTag; Stage: Fracture; Exact operand keys:id,tag,coordinates; Constitutional role: Compare values at two distinct sealed-record coordinates using one positive or counter tag | 0; Opcode:RequestHumanReview; Stage: Fracture; Exact operand keys:id,reason,members; Constitutional role: Emit lawful non-operative review matter over at least one sealed record | 0; Opcode:RejectUnlawfulAct; Stage: Fracture; Exact operand keys:id,rejected,fault; Constitutional role: Preserve anti-knowledge rejecting a 64-hex matter identity or act digest under a registered fault | 0; Opcode:AssertLayoutRelation; Stage: Forge; Exact operand… - 3. The Evidence Alphabet: 17 Tags and 12 Comparators: Subtopics: The 47 registered fault classes — EISA's tags are not natural-language hints. Each positive or counter tag binds to one deterministic comparator. The two inferential tags are reserved for transitive composition and are refused in
BindEvidenceTag. — Table Data: 1; Evidence tag:PHONE_STRONG; Class: Positive; Comparator and exact predicate:DigitsEqual: both digit strings have at least seven digits and are identical | 2; Evidence tag:PHONE_ONE_DIGIT_VARIANCE; Class: Positive; Comparator and exact predicate:DigitsOneSubstitution: equal-length digit strings, Hamming distance exactly one | 3; Evidence tag:… - 4. Sealed Cases: Two Substrates, One Artifact Identity: Subtopics: File cases; Passage cases — An act may point only into a sealed case. EISA creates
eisacase_<32hex>by hashing the Room name, a NUL separator, and the case's RFC 8785 JCS record digest. — Items: prepares and validates the citation for the caller and task; | refuses handles from another Room or unrevealed collaborative-pass findings; | loads and reverifies the canonical view and live source; | cuts the exact Unicode-scalar passage using the handle's selector; | refuses… - 5. The Four-Gate Verdict Machine: Subtopics: Counterevidence is submission-complete; Transitive closure is a path proof, not a slogan; Partition verification is global across the admitted submission; Consensus identity removes narrative, not responsibility — Each act ends in one consequence class: A malformed whole submission is
inadmissible. A recognized submission reportsdecision: partition,review,relation, ornone; it does not acquire a blanket database-execution verdict. ForAssertDuplicatePair, at least one admitted positiveBindEvidenceTagon the same pair is required. — Table Data: 1. Syntax and stage isolation; What the recognizer establishes: Strict JSON, exact keys, closed symbols, authenticated actor, unique IDs, bounded sizes, one act family, matchingartifactHash; What it refuses to let the proposal do: Smuggle duplicate fields, foreign opcodes, mixed stages, self-declared actors, or unbounded matter | 2. Spatial containment; What the recognizer establishes: Every coordinate resolves inside the named sealed case; every layout citation names the same committed view, revision, parser, and page; What it refuses to let the proposal do: Point at foreign records, manufactured geometry, another Room's evidence, or unstored fields | ****3. — Items:operative— a recognized act with a consensus identity. |lawful_non_operative— a valid review request, which creates no operative identity claim. |anti_knowledge— a valid rejection of other matter. |typed_fault— preserved residue with a fault code and no standing. | Exact disjoint cover. Every group member exists in the case, appears in at most one group, and every sealed record appears somewhere. Empty groups are meaningless. - Exhibit A — A Valid Contact Submission: The following synthetic submission binds two email relationships, asserts a pair, and proposes a complete partition over three sealed records.
artifactHashis the case's bare digest. - Exhibit B —
counterevidence_suppressed: Suppose the same submission also admittedact_bind_counter_03, bindingDOMAIN_ROOT_MISMATCHon the pair, butact_dup_101_102omitted it fromcounter. Its recognized-act report entry is: The refusal is not a score penalty. - 6. Proof Lane A: Spatial Relations Without Visual Guesswork: Subtopics: From PDF operators to integer evidence; The
CONNECTSproof; Branch labels enter relation identity — A flowchart can have perfect OCR and still produce a false graph. Recognizing the words inside two boxes does not establish which arrow connects them, which direction the arrow points, whether the tip touches a node or an enclosing frame,… — Items: Provenance. All three records come from one Room, view digest, source revision, parser digest, and page. Each record's committedLayoutFact.nodeis exactly the node its evidence handle cites. - Exhibit C — An Operative Layout Relation: This is the layout portion of a synthetic recognized-act report. The frame fields are observations in the current report shape, not an agent-authored
layout_proofobject. - Exhibit D — Two Distinct Partition Failures: The first splits records already proven equivalent. The second groups records without proving connectivity. They are different defects and receive different faults.
- 7. Proof Lane B: Semantic Switch First, Then the Entity-Resolution Stack: Subtopics: Stage 0 — Semantic Switch: protected computation before external AI; Stage 1 — qBlock: candidate nomination over protected records; Stage 2 — EISA Fracture: identity adjudication with dissent preserved; Stage 3 — EISA Forge: canonical state is a separate decision; Stage 4 — Shield: the baseline moves forward or does not move — Entity resolution adds a different danger: the evidence can contain direct identifiers, and the final merge can be difficult or impossible to reverse through the destination's ordinary product operations. — Table Data:
SS01: Privacy-first stage responsibility. Semantic Switch transforms native records before provider-exposed qBlock, Fracture, Forge, or Shield computation. Separate stages own identity judgment and destination mutation. — Items: every surviving value names its source record and field pointer; | provenance and validity intervals remain inspectable; | declared field quorums and canonical claims decide support; | unsupported values do not win through confidence alone; | a conditional reversal plan records… - 8. From Duplicate Records to Evidence-Linked Relationship Graphs: The commercial value is not “AI deduplication” as a slogan. It is a defensible identity substrate on which relationship paths can be inspected. A six-hop query can be a useful exploration horizon.
- 9. The 2026 Research Landscape: Proposal Quality Is Not Transition Standing: The current research supports a hybrid division of labor: models can improve nomination, comparison, and review, while deterministic systems must decide which declared predicates actually hold. — Table Data: Teofili et al., Can we trust LLM Self-Explanations for Entity Resolution?, June 2026 preprint; What it establishes: Across three LLMs, ten datasets, and multiple prompting strategies, self-explanations were often unstable, weakly faithful, and poorly aligned with counterfactual evidence.
- 10. The Continuous Wire into the Room: EISA recognition and Room admission share one path.
recognizepreviews a verdict;submitruns the same recognition logic and records the result. — Items:byand nativesession; |roomand optionaltask; | the case's bareartifactHash; | the complete recognition result; | submission time. - Exhibit E — The Current Room Ledger Envelope: This synthetic
showresponse illustrates the current ledger shape with one operative binding. The per-act report remains insideresult. The immutable node is constitutional proof matter for the next boundary. - Bring One Bounded Protected-Computation Problem: The Semantic Switch & Entity Resolution Challenge applies this constitutional layer to one bounded contact, company, product, or other duplicate-record workflow.
- The Article 03 Claim Card: Items: Mechanism: | Semantic Switch Protected Computation + EISA Constitutional Act Recognition | (native-data minimization, relation-isomorphic fictional records, | qBlock candidates, Fracture, Forge, local exact restoration, Shield, | strict parsing, sealed cases, pointer law, deterministic comparators, | Four-Gate recognition, layout proof, and Room ledger) | Failure it governs: | Native identifiers exposed to external AI computation; | comparison structure destroyed by destructive redaction; | ambiguous or…
- Before the Room Can Govern, the Act Must Mean One Thing: Return to the three agents voting for a merger. The conventional system can report that the vote was unanimous, the JSON matched a schema, and the judge was 97% confident.
- Frequently Asked Questions: Subtopics: Why is JSON Schema or constrained decoding insufficient to govern AI actions?; How does EISA prevent an AI agent from suppressing conflicting evidence?; Does
consensusIdentityprove independent agreement?; Why does Semantic Switch run before qBlock?; Is Semantic Switch anonymization?; How does EISA verify flowchart connections without trusting VLM prose?; Does an operative EISA verdict execute the proposed transition? — Constrained decoding establishes shape. It does not resolve record-relative pointers against a sealed case, run deterministic comparators on the values, reject faulted dependencies, force counterevidence into the decision, or prove that a… - Continue the Series: Items: Series index: From Proposal to Consequence: The AINA Transition-Governance Series | Previous: Article 02 — Zero-Trust Document Ingestion and Server-Cut Evidence…
- Primary 2026 Sources: Items: Teofili et al. — Can we trust LLM Self-Explanations for Entity Resolution?,
arXiv:2606.01210| Papadakis et al. — AgenticER: the next frontier in Entity Resolution,arXiv:2607.27435| [Smith et al. - Publication Metadata: Title tag: Semantic Switch and EISA: Protected Computation and Entity Resolution Meta description: Semantic Switch protects native data before external AI computation, then qBlock, Fracture, Forge, and Shield turn fictionalized…
FAST 3-STEP SERIES DIGEST 1/3 — Digest 1 (§01–§03) · Digest 2 (§04–§06) · Digest 3 (§07–§09) · 1-Page Master Brief · All 9 Articles · Assurance · Proof