Domain DAT

Information people can trust

Start here · Questions · Progress · People decide

Keep the source, time, authority, permissions and corrections with information people act on.

Question	What we are working out	AI-generated potential answers and research	Work state
DAT-01	Which official and service-information owners should supply each essential information type, and how should updates reach the network?	AI-generated option: Create a source agreement for every essential information type, beginning with Redland's official disaster information and the actual transport, utility and service owners. Record update rights, checking intervals, failed checks and the approved offline copy. AI-generated future concept: Give each Aura a source-aware explanation layer that can show the original message, its age and applicable area. Future automated source checks could propose corrections or escalation while issuers retain control of their warnings. Research examples: Redland City Council: disasters and emergencies · Sunshine Coast: Disaster Hub helps Coast stay safe	open
DAT-02	What information should each record carry so people can understand its source, timing, current meaning and later corrections?	AI-generated option: Give each record a stable identity, source, issue and receipt times, applicable place, status and replacement links. Show information age and clock confidence plainly, and retain corrections so people can understand what changed. AI-generated future concept: A personal Aura could explain a record's history and why an older edition is no longer current. XR views could show the same provenance beside a map, with the ordinary text record always accessible. Research examples: Common Alerting Protocol Version 1.2 · Managing records created using artificial intelligence	open
DAT-03	Who should prepare, review, approve, prioritise and withdraw each kind of public information?	AI-generated option: Use a publishing table that names who prepares, checks, approves, withdraws and corrects each information category. Bind every delegation to its scope and edition, with recoverable access and a visible record of accountable overrides. AI-generated future concept: Auras and advanced AI/SI could prepare alternate formats and identify missing sources, while a permissions service restricts publication to the exact approved content and scope. Urgent delegated actions could be pre-agreed and reviewed afterwards from their full logs. Research examples: Australia's AI Ethics Principles · Common Alerting Protocol Version 1.2	open
DAT-04	How should records come back together after locations reconnect, keeping corrections, cancellations and outstanding work clear?	AI-generated option: Keep an append-only change history so disconnected updates, cancellations and requests can be brought together without silently losing meaning. Automatically combine compatible changes and present genuine conflicts to the responsible person with their source histories. AI-generated future concept: Future Auras could prepare a readable conflict comparison and propose a resolution grounded in the signed records. A twin could replay reconnection sequences to expose double allocations before software is used with real supplies or requests. Research examples: Common Alerting Protocol Version 1.2 · Managing records created using artificial intelligence	open
DAT-05	What information should a person choose to share for help or personal AI guidance, and how should they control it afterwards?	AI-generated option: Collect the least information needed for the agreed help task, keep optional Aura context under the person's control and separate public records from private casework. Explain recipients, retention, correction, export and deletion before sharing. AI-generated future concept: Explore a personal Aura-held profile that releases only the selected information for a particular task, with expiring permissions and a readable sharing receipt. Test how withdrawal affects cached copies, backups and already-completed actions. Research examples: Privacy and the use of commercially available AI products	open
DAT-06	How should local reports, media, event changes and accessible versions move from contribution to reviewed publication?	AI-generated option: Offer a contribution-to-publication path with source type, permission, accessible versions, named review and withdrawal. Keep unconfirmed reports clearly labelled, preserve an original alongside any AI adaptation and let contributors see the response to their input. AI-generated future concept: An Aura could draft a translation, easy-read version or XR presentation with links back to its source and human review. Cultural custodians could set audience and reuse permissions that every format and downstream assistant carries forward. Research examples: AIATSIS Research Ethics Framework · Privacy and the use of commercially available AI products	open