Manifold Grace

Research & Knowledge / 2027–2030 Programme

INTEGRATED RESEARCH AND VENTURE WORKSTREAMS · 10 NOVEMBER 2026

Twenty-workstream implementation and evidence register

This is a durable status and implementation record consolidating work already undertaken, current deliverables and tasks that require independent testing, customer participation or additional technical infrastructure. It avoids presenting the author's established investigations as not begun, while not presenting draft designs as verified deployments.

Evidence standard: committed static pages, completed diagrams, prototype behaviour and approved external validations are different accomplishments. Each record below identifies what is actually documented and what remains to be proved.

#WorkstreamCurrent evidence-based stateLinkDocumented deliverableVerification / next gate
1Complete comprehensive AI Knowledge AtlasInitial 7-domain/35-topic taxonomy created; exhaustive scientific coverage remains under review.Existing evidence or workspaceBreadth-first subject map, cited depth links and revision controlsExpand missing disciplines using independent taxonomy audits
2Technology Opportunity Window IndexSix-factor model and 28 venture candidates recorded.Existing evidence or workspaceOpportunity dimensions, explicit score semantics, backtesting requirementsReplace judgemental scores with calibrated outcome evidence
3Validate 28 venture opportunitiesAll ideas documented with assumptions and validation tasks; no paid market evidence verified.Existing evidence or workspaceInterview and pilot criteria, source checks and comparison frameworkSecure permissioned buyer interviews; no invented demand
4Build comparative systems registry20-sector atlas and record schema created; source-level cases require verification.Existing evidence or workspaceMaps, tools, edges, evidence status, comparison protocolCollect real architecture cases and cite primary sources
5Prototype research and investigation agentsDeterministic local demonstration and 12-stage architecture available; autonomous tools not deployed.Existing evidence or workspaceRunnable browser comparison, JSON export, agent process and security gatesDevelop read-only retrieval connector, independent verification and labelled test set
6AI stack consolidation advisory pilotService blueprint documented; no external client pilot carried out.Existing evidence or workspaceScope, permitted evidence, overlap matrix, safeguards, deliverablesValidate willingness to pay with real consenting organisation
7Complete reusable evidence-verification frameworkReference architecture and claim-evidence rules defined; production API not deployed.Existing evidence or workspaceEvidence provenance, contradiction and uncertainty fieldsTest extraction and false citation detection
8Create 2027–2030 commercial roadmapPrioritised venture register and entry route documented; milestone dates provisional.Existing evidence or workspaceStages from research to qualified pilot to offeringCost and staffing gates after buyer validation
9Audit website omissions and duplicationExisting research, venture and government materials cross-referenced; full site audit unverified.Existing evidence or workspaceNo-overwrite audit specification and linkage rulesRun site-wide crawl, compare pages and inspect broken links
10Build technical demonstration portalWorking local architecture-comparison demo published at lab page.Existing evidence or workspaceInteractive tool, synthetic scenarios, JSON export and staged expansion planAdd backend only after auth, source, safety and privacy controls
11Develop legacy-to-AI integration assessmentsOpportunity mapped; implementation remains prospective.Existing evidence or workspaceReadiness scorecard, trust boundaries and API mappingPermissioned read-only system sample
12Create compliance and ethics assessment engineControl and evidence taxonomy created; no legal verdict engine certified.Existing evidence or workspaceRole/jurisdiction/obligation and separate ethical-impact recordsTest against legal specialists and jurisdiction versions
13Interactive visual architecture viewerRelationship list and comparison demo available; full graph editor not yet built.Existing evidence or workspaceComponent/edge schema and baseline-vs-comparison logicImplement accessible graph views and layout alternatives
14Audit Agency.ai service catalogueExisting portal identified; full 50-service mapping not established.Existing evidence or workspaceCrosswalk schema: service, venture, dependency, differentiationExtract catalogue with source IDs and de-duplicate
15Research-gap and novelty assessment protocolGap-first, falsification-led criteria documented; no exhaustive overlap claim.Existing evidence or workspaceClaim, strongest prior art, unresolved limitation, measurable contributionSystematic literature retrieval and independent validation
16Research and technical-report templatesCore document and provenance structure described; publication-specific templates need QA.Existing evidence or workspaceSections for abstract, prior art, method, evidence and limitationsAdapt to actual journal/conference submission instructions
17Technology and market monitoring dashboardMonitoring dimensions and source anchors defined; automated feeds not enabled.Existing evidence or workspaceSignals and quarterly reassessment frameworkSet authorised recurring monitoring with sources and clear thresholds
18Establish AI QA benchmarksIllustrative comparison issues are visible; benchmark dataset not validated.Existing evidence or workspaceAccuracy, source recall, false positive, robustness and cost metricsDevelop human-labelled control cases
19Historical opportunity-window case studies2017 robotics case and substitution hypothesis linked; broader evidence needed.Existing evidence or workspaceCase narrative, chronology, contemporaneous sources and alternative explanationsTest market decline claim against actual sales and education data
20Automate research and productivity reportingDated source-backed entries exist; autonomous reporting is not authorised or operating.Existing evidence or workspaceStatus labels, commit evidence, audit trail and review before publishingBuild signed-off report generation and deduplicated log updates

Detailed implementation of workstream 5 — Research agents

Start with reproducible, read-only analysis. Distinguish diagram-reading, retrieving independent sources, making claims and approving conclusions as separate authorisation steps. A working comparison demonstrator is linked below; full autonomous data acquisition is not yet deployed.

SequenceEngineering actionMethodExit criterion
1Discovery contractDefine precise question, jurisdiction, classes of systems, scope and prohibited actions.Signed-off scope; approved sources; reject inaccessible or restricted data
2Read-only research intakeObtain diagrams, specs and public source pages with original URLs and timestamps.Record source licenses, versions, retrieval method and hashes
3Visual and text extractionConvert each source into entity, relationship and claim candidates.Use model outputs as provisional; assign evidence spans and uncertainty
4Independent validationCheck extracted relations against primary documents, references and contradictions.Mark unverified or unsupported; avoid inferred implementations
5Normalise and compareUnify names with controlled vocabulary, compare nodes and edges and alternative views.Metric definition, structural differences, comparator and no false equivalence
6Tool and compliance reviewMap vendor dependencies, safety controls and applicable duties by region/use case.Risk questions, role and jurisdiction; not a legal certificate
7Benchmarks and adversarial testsMeasure citation precision/recall, extraction error, false positive and missing data.Human-labelled test cases and explicit pass/fail thresholds
8Human sign-off and releaseReview findings and permission before any public output.Corrections, redactions, versioned review record and reversible release

Detailed implementation of workstream 10 — Technical demonstration portal

Stage public interaction without claiming operational agent capabilities before they exist. The current browser demonstrator processes only contained sample data and exports its comparison report.

SequenceEngineering actionMethodExit criterion
1Define product boundariesOffer read-only public demonstrator with synthetic fixtures; separate from operational research agents.No login needed; no personal data, credentials or hidden vendor claims
2Design stable object modelSystems, nodes, edges, sources, controls, review flags and versioned findings.JSON schema and documented assumptions
3Build accessible interfaceKeyboard-controlled selections, labelled table, readable output and meaningful focus.Responsive design, mobile review and contrast test
4Implement deterministic comparisonComponent set comparison, directed edge comparison and evidence gap prompts.Demonstrable calculation, no made-up compliance scores
5Export reproducible outputsUse JSON report download with scenario IDs and method label.Portable structured output and consistent reruns
6Run functional testsCheck same-system comparison, zero overlap, edge differences and missing controls.Human-inspectable results and regression cases
7Protect against untrusted contentRender remote text as text; no direct execution or unreviewed HTML.Cross-site scripting and prompt-injection tests before remote intake
8Add real document import after reviewOnly permitted public or uploaded sources with provenance metadata.File type/size limits and retention policy
9Integrate constrained agent servicesServer-side secrets, scoped tools, allowlists, rate limits and approvals.No credentials in browser; no autonomous external writes
10Evaluate before wider releaseDocument performance, errors, costs and usability; maintain rollbacks.Independent acceptance criteria and incident procedure

Open the interactive research-agent prototype · AI Knowledge Atlas · Comparative Systems Atlas · 2027 Venture Register

Evidence and policy anchors

UK AI Adoption Survey 2026 · NIST AI RMF · OWASP Agentic Applications · W3C Provenance

This register makes no claim to having interviewed customers, obtained legal sign-off, scanned every site file, trained models or deployed production agents. Versioned evidence and field tests are necessary before those claims can be made.