Case study 04 · Three-report research series

AI Help, Living Well and Democratic Renewal

Can Australia's existing systems manage interacting risks to 2035, where is AI necessary or useful, and can broad AI-enabled civic resilience be delivered safely and durably?

What is being tested: A three-stage test of the strongest feasible non-AI comparator, AI necessity and human benefit, and the delivery, safeguards and endurance of broad AI-enabled civic resilience.

Cumulative portfolio verdict

UNRESOLVED

No function is demonstrated NECESSARY. Anticipation and forecasting, and cyber and AI-threat management, are USEFUL NOT NECESSARY. Climate, infrastructure and logistics; care and living well; information judgement; and deliberation remain UNRESOLVED. P2 is the most credible near-term delivery pattern but remains partial; P3 and P4 remain unproven.

Boundary: This is not a recommendation, forecast or claim that AI is either required generally or useless. It preserves a strong non-AI comparator and keeps missing delivery, safeguard, ecological, distributional and democratic evidence visible.

Three-stage test

The argument is tested in sequence

  1. 1
    Stage 1

    Whether Australia's existing systems and strongest feasible non-AI reforms are adequate.

  2. 2
    Stage 2

    Whether AI is necessary or only useful, and whether it helps Australians live and decide well.

  3. 3
    Stage 3

    Whether broad AI-enabled civic resilience can be delivered, safeguarded and sustained.

Function results

Usefulness is not necessity

No function is demonstrated necessary.

FunctionResult
Anticipation and forecastingUSEFUL NOT NECESSARY
Climate, infrastructure and logisticsUNRESOLVED
Care and living wellUNRESOLVED
Cyber and AI-threat managementUSEFUL NOT NECESSARY
Information judgementUNRESOLVED
DeliberationUNRESOLVED

Pathways: P2 is the most credible near-term delivery pattern but remains partial. P3 and P4 remain unproven. No pathway passes the whole-system test.

Accepted reports

Read the full evidence in order

Report 1 — Can Australia’s Existing Systems Manage the Interacting Risks to 2035?

COMPARATOR GAP, NOT AI NECESSITY. The launch baseline plus the strongest feasible non-AI reform package is not demonstrated adequate as a whole. That comparator gap does not establish that AI is necessary.

Read Report 1 in full

Report 2 — Where Is AI Necessary and Can It Help Australians Live and Decide Well?

NO FUNCTION DEMONSTRATED NECESSARY. Anticipation and forecasting, and cyber and AI-driven threat management, are useful but not necessary. The other four functions and every function's mandatory gate state remain unresolved.

Read Report 2 in full

Report 3 — Can Broad AI-Enabled Civic Resilience Work by 2035?

UNRESOLVED. Broad AI-enabled civic resilience is not yet shown to be deliverable, safeguarded and sustainable by 2035. P2 is the clearest bounded delivery case, but no pathway passes the whole system.

Read Report 3 in full

Challenge or extend the evidence

Contribute to the public record

Suggest a correction · Submit new evidence · Propose research

AI assisted the research and drafting under human Product Owner direction. Source inclusion does not imply endorsement by a government, regulator, author or institution.