Fictional cross-domain exercises
The casebook.
Test the same commitments where the stakes change. Every case uses synthetic details and supports tabletop learning—not live operations or domain authority.
Current coverage
Life First defines Tier 0 through Tier 3. The case labels shown below preserve the source repository's current classifications; its published exercises are presently labeled Tier 2 or Tier 3. See all four definitions in the Framework.
Community and human support
Protect choice, access, privacy, and qualified human authority.
Community resource referral
Can an advisory shortlist preserve choice without becoming an eligibility decision?
Read the case ↗ Tier 3Human-led mental-health support intake
Can an assistant support a qualified human without diagnosing, treating, classifying crisis, or initiating coercive action?
Read the case ↗Belief and pluralism
Support logistics without governing belief.
Pluralistic religious-community support
Improve logistics without profiling belief, ranking doctrine, or coercing participation.
Read the case ↗ Tier 3Sacred-gathering readiness
Support authorized humans with crowd, weather, and accessibility information without turning belief or attendance into a threat signal.
Read the case ↗Security and engineered systems
AI remains advisory where authority and protection must remain independent.
National-security watchfloor
Organize uncertainty without turning model output into intelligence, targeting, force authorization, or rights deprivation.
Read the case ↗ Tier 3Hazardous machinery
Keep AI support separate from engineered protection, interlocks, safe-state controls, and qualified authority.
Read the case ↗Remote Earth environments
Communication gaps do not expand AI authority.
Antarctic research station
Life, rescue capacity, environmental protection, crew agency, and fallback outrank schedule and research output.
Read the case ↗ Tier 3Arctic field expedition
Keep local stop authority clear when weather, shelter, communications, recovery, and outside support are uncertain.
Read the case ↗Remote space environments
Preserve human mission authority and independent life-sustaining controls.
These fictional cases illustrate delayed review, constrained repair, future fallback capacity, and high-consequence uncertainty. They do not describe real missions or procedures.
Orbital-station life support
Independent systems and crew authority remain primary during a fictional environmental anomaly.
Read the case ↗ Tier 3Lunar habitat
Protect survival loads, crew agency, repair margin, and abort or fallback capacity under resource constraints.
Read the case ↗ Tier 3Crewed Mars mission
Require corroborated evidence and independent authority before an irreversible mission phase.
Read the case ↗Casebook boundary
Examples are not operating instructions.
None authorizes AI to decide theology, diagnose or treat a person, authorize force, identify a target, override engineered protection, command a vehicle, or approve a mission phase. Qualified authorities and current domain rules remain controlling.