complete pilots built first
CURATION DECISION · NOT A NEW CONTENT QUEUE
Seventeen candidates. Two pilots. One method to prove.
The audit categories measured different things. This plan turns them into one operating hierarchy and gives every build a destination. No build is deleted. No build receives a carousel merely because it has a number.
LOCKED NOW
Canva stays frozen.
other public candidates held behind evidence gates
builds receiving no new standalone slides yet
builds deleted or misrepresented as production AI
TIER 1 · BUILD NOW
Consensus launch pilots
Consensus hero: human authority, evidence and stopping power are central to the RN method.
Consensus hero: inclusive repair is legible, interactive and directly connected to design practice.
- 01One precise question the prototype tests
- 02Exact users, actors and authority boundaries
- 03Primary sources and material counterevidence
- 04What the current code actually does
- 05Inputs, state changes, rules, outputs and failure paths
- 06What is synthetic, hard-coded, missing or prohibited
- 07Observed first-time-user testing and misunderstandings
- 08A literal materials-and-steps reconstruction guide
- 09Tests, accessibility checks and claim boundaries
- 10What a production version would require
- 11Three creative treatments before selecting a design system
- 12A release decision: publish, revise or stop
THE OTHER TWO INTELLECTUAL LEADERS
Method anchors, not launch episodes.
Builds 012 and 014 are intellectually strong and source-linked, but the external-value review did not find them distinctive enough to lead the public series. Their substance will strengthen the two pilots before either competes for its own release.
Supply the consequence, control and risk-tier architecture inside Build 001.
Supply the legal-judgment and nondelegable-authority analysis inside Build 001.
THE COMPLETE 17-CANDIDATE QUEUE
Compete for release; do not assume release.
Consensus hero: human authority, evidence and stopping power are central to the RN method.
Consensus hero: inclusive repair is legible, interactive and directly connected to design practice.
Place-sensitive transfer connects Hawaiʻi, implementation and Dutch field research.
Institutional handoffs are distinctive, but the prototype needs real workflow evidence.
AI vendor-claim scrutiny has immediate professional relevance and a clear evidence test.
Cannabis access fits RN’s domain authority but requires real journey evidence and current sources.
Island dependency is highly distinctive but must be grounded in actual infrastructure evidence.
The lived experience of rules is a strong RN proposition that needs observed human evidence.
Citation verification directly connects law and AI, but must be tested against real citations.
Implementation after enactment is a strong legal-systems story that needs a real law and timeline.
A powerful capstone if presented as a bounded planning rehearsal—not a forecast or digital twin.
Strong public explanation concept; prove truth-preserving translation with one real claim.
Strong discovery method; prove that it finds a real unserved decision.
Useful product judgment; show one decision where refusing software was the correct result.
Promising knowledge infrastructure; demonstrate a real source changing a claim and output.
Strong link to RN’s reporting practice; run it on one rights-cleared reporting packet.
Strong portfolio mechanism; prove confidentiality and rights gates using one real cleared project.
THE 53 SUPPORT-OR-MERGE RECOMMENDATIONS
Modules inside larger stories.
These builds retain code, routes, provenance and build numbers. They lose the presumption of a standalone episode. The groups below cover all 53 intellectual support-or-merge recommendations.
EXPLICIT MERGE GROUP · LEAD 001
Stop, escalate and release consequential AI
Combine escalation, evaluation, tracing and release readiness into the human-authority story.
EXPLICIT MERGE GROUP · LEAD 063
Remember why a decision was made
Combine evidence ledgers, decision memory and institutional knowledge into one living-research system.
EXPLICIT MERGE GROUP · LEAD 030
Minimize, permit and understand product data
Treat consent, minimization, data quality and privacy behavior as parts of inclusive system repair.
EXPLICIT MERGE GROUP · LEAD 001
Give AI bounded work and governed memory
Collapse generic AI-team infrastructure into one inspectable authority, orchestration and proof architecture.
SUPPORT CLUSTER · LEAD 003
Discover the real decision
Methods and stress tests supporting discovery, ownership, readiness and institutional-friction analysis.
SUPPORT CLUSTER · LEAD 030
Design for access, place and embodiment
Portability, adaptive environments, personal context, visible personalization and equity become supporting lenses.
SUPPORT CLUSTER · LEAD 063
Build evidence that stays alive
Retrieval, lifecycle, comparison, consequence and translation become modules in a source-linked evidence system.
SUPPORT CLUSTER · LEAD 080
Turn work into public infrastructure
Audience signals, content experiments, media operations, relationships and founder memory support the reporting-to-infrastructure story.
SUPPORT CLUSTER · LEAD 045
Ground access in lived experience
Synthetic access comparisons and continuity rehearsals remain modules until field evidence exists.
53/53 mapped. No support-or-merge record is orphaned.
TEN EXPLICIT CUTS
Cut the standalone episode—not the work.
Escalation is not strong enough to carry a separate episode.
The evidence ledger becomes the memory/provenance layer.
Minimization belongs inside the inclusive data-behavior system.
Data cleaning is an implementation module, not a flagship story.
Privacy behavior becomes a repair dimension.
Domain evaluation supports the release-and-escalation architecture.
Agent tracing becomes evidence for human review.
Job definition belongs inside bounded AI authority.
Team design alone is too generic for a standalone episode.
Orchestration becomes an implementation layer, not a headline.
REMAINING BUILDS
Preserve, repair or use as evidence.
Other retained or repair builds
Useful supporting modules and promising incomplete builds that were not among the intellectual audit’s 53 support-or-merge records.
Archive concepts
Keep clearly labeled as synthetic thought experiments until field evidence changes their status.
RELEASE RULE
A build advances only when the evidence advances.
No fixed quota. No promise that all seventeen ship. The two pilots are built and tested first. Every later candidate must outperform them on truthfulness, technical clarity, RN distinctiveness, child-clear reproducibility, accessibility and visual force.