2 · CARE & FAIR Data Handling Lifecycle
Explore all 8 stages from planning to destruction & assurance
Select any lifecycle stage below (use mouse, touch, or Left/Right arrow keys) to inspect
its proposed controls, CARE and FAIR alignment, and operational KPIs. Every control ID
links directly to its full ISM-style specification in the manual.
Establish community authority above organisational oversight, embed Aboriginal and Torres Strait Islander participation in data governance bodies, define separation of duties, and capture community data priorities and capability support needs before collection begins.
Proposed Controls (7)
-
Community authority over Indigenous data
-
Governing body oversight below community authority
-
Aboriginal and Torres Strait Islander people in data governance
-
Defined roles and separation of duties
-
Custodian appointments and revocations come from the community
-
Support for community data capability, on the community's terms
-
Community data priorities
How This Stage Is Measured
KPI-LC1 Datasets with nominated community authority
(Datasets with named community authority / Total Indigenous datasets) x 100%
- ML1
- Register exists
- ML2
- 100% mapped before use
- ML3
- 100% system-enforced
KPI-LC1b Community data priorities actioned & reported
(Documented priorities reviewed & reported on schedule / Total recorded priorities) x 100%
- ML1
- Priorities requested
- ML2
- >= 90% reviewed on cycle
- ML3
- 100% tracked with community sign-off
Obtain and record free, prior and informed consent (FPIC) collectively before ingesting or creating data, uphold the right to decline, vary or withdraw consent, suspend processing immediately when new information arises, and review data stewardship and quality with the community.
Proposed Controls (5)
-
Right to decline and to withdraw
-
Free, prior and informed consent before collection
-
Consent variation and withdrawal
-
Suspension when new information comes to light
-
Data stewardship and data quality reviewed with the community
How This Stage Is Measured
KPI-LC2 Ingestion batches gated by valid collective FPIC
(Ingested Indigenous batches with valid FPIC record / Total batches) x 100%
- ML1
- Manual consent file kept
- ML2
- 100% system-checked on ingest
- ML3
- 100% verified against community register
KPI-LC2b Consent withdrawal / new-information suspension latency
Time from recorded withdrawal or new-info flag to access refusal
- ML1
- Within agreed SLA
- ML2
- Same day (<= 24 hours)
- ML3
- Immediate (automated fail-closed)
Attach ordered sensitivity classifications that automated systems can only raise, bind cultural protocols and Local Contexts TK/BC labels directly to every record so they travel across copies and derivatives, fail closed on existence hiding for restricted records, and maintain protocol-aware FAIR metadata.
Proposed Controls (3)
-
Every record carries a classification that can only be raised automatically
-
Cultural protocols and Local Contexts labels travel with the data
-
Existence of restricted records fails closed
How This Stage Is Measured
KPI-LC3 Classified & protocol-labelled record completeness
(Records with valid classification + applicable protocols/labels / Total records) x 100%
- ML1
- Scheme & defaults set
- ML2
- 100% classified; strictest on merge
- ML3
- 100% raise-only enforced; synced labels
KPI-LC3b Existence-hiding leak rate across read paths
Detected leaks of hidden records across search, counts, maps, APIs and errors
- ML1
- Excluded from main lists
- ML2
- 0 leaks across read paths
- ML3
- 0 leaks verified automatically per change
Protect sensitive data at rest with application-level envelope encryption and in transit with verified TLS, isolate tenants at the database engine layer, keep data onshore by default, disclose sub-processors, match recipient environments to data sensitivity, and offer key custody up to community-held keys.
Proposed Controls (8)
-
A secure environment matched to the data
-
Key custodians
-
Encryption at rest for sensitive Indigenous data
-
Encryption in transit
-
Key custody options up to community-held keys
-
Tenant isolation enforced by the database
-
Data residency onshore by default
-
Sub-processors are disclosed
How This Stage Is Measured
KPI-LC4 Sensitive payloads sealed with per-custodian-group keys
(Sensitive+ payloads envelope-encrypted under group/tenant key / Total sensitive+ payloads) x 100%
- ML1
- Storage encryption active
- ML2
- 100% app-level sealed incl. history
- ML3
- 100% verified plaintext-free; community key option
KPI-LC4b Onshore residency & sub-processor disclosure compliance
Unapproved cross-border deployments or undisclosed sub-processors
- ML1
- Hosting region & list documented
- ML2
- 0 off-region deploys (IaC guarded)
- ML3
- 100% runtime & AI services jurisdiction-checked
Prevent operator roles and break-glass accounts from overriding cultural protocols, require strong MFA and step-up authentication, gate joiner/mover/leaver access on vetting and cultural competency training, enforce purpose limitation, require fresh approval for secondary uses and derivatives, supply disaggregated data on community request, and ban unauthorised AI training or prompting.
Proposed Controls (19)
-
A named accountable officer
-
Only approved users
-
Use only for the agreed purpose
-
No AI training or processing without agreement
-
Recipient staff are trained before data is released
-
Operator roles cannot override community authority
-
Cultural competency and data sovereignty training before access
-
Vetting and suitability
-
Strong authentication for privileged and custodian roles
-
Joiners get the least access they need, once approved
-
Movers have their access reviewed
-
Leavers and project ends remove access promptly
-
Outside parties have time-limited, purpose-limited access
-
Purpose limitation
-
Approval for secondary use and derived data
-
Periodic access review by the custodian group
-
Disaggregated data for the community on request
-
Access control that operator roles cannot override
-
AI and model training restrictions
How This Stage Is Measured
KPI-LC5 Protocol-bound operator override refusal rate
(Refused operator/break-glass/AI attempts on protocol-bound data / Total attempts) x 100%
- ML1
- Policy prohibits override
- ML2
- 100% refused by central authz
- ML3
- 100% verified per change & visible to community
KPI-LC5b Same-day leaver/revocation removal & training currency
(Access removed <= 24h & active users with current cultural training / Total) x 100%
- ML1
- Checklist & training run
- ML2
- 100% same-day removal; dates tracked
- ML3
- Automated SCIM deprovision + lapsed-training block
Release data only under custodian-signed data sharing agreements with restrictive defaults, attach classification, protocols and labels to every export, apply sacred-site location generalisation and deceased-person presentation controls, prohibit onward sharing and re-identification, mandate community review before publication, and deliver agreed benefits and attribution.
Proposed Controls (10)
-
No onward sharing
-
No re-identification and no unapproved linkage
-
Community review before publication
-
Benefit sharing and attribution
-
Data sharing agreements with required clauses
-
Agreement review and expiry
-
Onward sharing is controlled by the sharing organisation
-
Re-identification risk is controlled before release
-
Exports only under an agreement, with protocols attached
-
Presentation controls on display and export
How This Stage Is Measured
KPI-LC6 Exports governed by active custodian-signed agreements
(Exports linked to active agreement with intact protocols / Total external exports) x 100%
- ML1
- Manual approval & log
- ML2
- 100% carry protocols & watermarks
- ML3
- 100% system-gated by agreement, scope & format
KPI-LC6b Pre-publication community review & benefit delivery
(Outputs reviewed before release & agreed benefits reported / Total outputs) x 100%
- ML1
- Clauses in every agreement
- ML2
- 100% sent for review & logged
- ML3
- Community co-interprets & verifies benefit delivery
Enforce agreement-bound retention limits with verified purges across primary stores and backups, enable communities to repatriate data in usable formats with protocols intact, support cryptographic erasure per custodian group, require signed destruction attestations, and notify communities promptly of any breach or cultural harm.
Proposed Controls (7)
-
Tell the community about any breach
-
Destroy or return the data at the end, and attest to it
-
Incident and breach notification to the community
-
Destruction and attestation of destruction
-
Return of data to the community
-
Cryptographic erasure per custodian group
-
Retention limits and verified purge, including backups
How This Stage Is Measured
KPI-LC7 Timely destruction & signed attestation completion
(Expired/withdrawn datasets destroyed with signed attestation / Total due) x 100%
- ML1
- Procedure covers all copies
- ML2
- 100% attested & delivered on time
- ML3
- Key destroyed, chain-logged & verified post-restore
KPI-LC7b Community breach notification SLA adherence
(Incidents notified within agreement SLA or <= 72h & never after regulators / Total) x 100%
- ML1
- Contacts & SLAs documented
- ML2
- 100% logged & procedure tested
- ML3
- Detection feeds notice; community co-reviews
Operate a tamper-evident hash-chained audit log that communities can independently verify, perform independent access audits, maintain protocol-aware community data inventories, run scheduled CARE/FAIR self-assessments with community review, anchor signed automated evidence bundles to the audit chain, verify transfer logs and backup restore drills, and commission impartial partnership health checks.
Proposed Controls (9)
-
Regular CARE and FAIR maturity self-assessment with community review
-
Continuous automated control evidence anchored to the audit chain
-
Community data inventory and FAIR metadata quality verification
-
Periodic verification of data transfer logs and backup restoration drills
-
Independent partnership and governance health check
-
Access records available to the community
-
Independent audit of access, visible to the community
-
Tamper-evident audit log
-
Communities can verify the audit record independently
How This Stage Is Measured
KPI-LC8 IDG Domain & Overall Maturity Level (IDG-ML0 to ML3)
Domain = min(applicable controls); Overall = min(applicable domains)
- ML1
- IDG-ML1 across all domains
- ML2
- IDG-ML2 + community review active
- ML3
- IDG-ML3 + automated evidence & health checks
KPI-LC8b Automated evidence pass rate & audit chain integrity
(Passing mapped checks / Total mapped checks) x 100% with 0 chain breaks
- ML1
- Checks run pre-release
- ML2
- 100% pass; Ed25519 bundle anchored
- ML3
- Verified with pinned key; monthly transfer & annual restore drills
Existence hiding over open indexing (FAIR F4)
Where a community protocol requires it, unauthorised people cannot learn a restricted record exists in search, counts, maps or error messages (IDG-TEC-003).
Cryptographic erasure over persistent metadata (FAIR A2)
Communities can purge data and destroy per-group encryption keys on withdrawal or expiry; only the signed destruction attestation remains (IDG-TEC-007, IDG-TEC-016).
Fresh community consent over generic licences (FAIR R1.1)
Consent never launders into derived outputs or AI models. Every secondary use and derived dataset requires fresh community approval (IDG-PRC-004, IDG-TEC-017).