Ethics Readiness Monitor
Leveraging the GC-REAIM Strategic Guidance Report, this Watchdog grades current AI‑military developments and surfaces compliance signals by region and AIMID category, with top scorecards and rationale highlights.
Why this matters: The Ethics Tracker offers a directional view of ethical maturity across world regions and AI-military sectors based on scored reporting.
Disclaimer: This monitor is not endorsed by REAIM. It was created using the GC-REAIM report as its theoretical foundation.
Methodology
How to interpret the Ethics Tracker
A concise guide to what the scores represent and how they should be read.
- Scope: This dashboard summarizes AIMID articles from the specific source_type subsets currently scored (Academic Journal, Defense Publication, Government Contract). It reflects only the subset with completed scorecards and should be read as coverage of scored content, not total reporting volume.
- Scoring logic: Each article is evaluated across 10 GC-REAIM dimensions. Every dimension is scored 0 or 1 based only on explicit mentions in the text. The overall score is computed as (sum of dimension scores / 10) * 5, and the letter grade mirrors that score.
- Evidence: When a dimension is mentioned, the scorecard lists 1-3 supporting snippets and a one-sentence rationale. If a dimension is not discussed, the evidence is marked N/A.
- Aggregation: The dimension-by-region chart shows the average percent of scored articles mentioning each dimension. The category chart shows average scores by AIMID category. The top scores list surfaces the highest scoring scorecards with their rationale excerpts.
- Purpose & maturity: The intent is to build a picture of ethical maturity across world regions and AI-military sectors by comparing scores along the same dimensions. This is an early-stage product and not a finalized assessment.
- Source & interpretation: Scoring criteria are drawn from the REAIM Strategic Guidance Report. Use these results for directional comparison; gaps indicate no scored coverage yet.
-
Scoring dimensions (10)
- International Law & Ethical Foundations: References to legal or ethical obligations such as IHL, human rights, proportionality, distinction, accountability, civilian protection, or dignity.
- Human Agency, Control, and Accountability: Clear human decision authority and responsibility for consequential or targeting decisions.
- Risk Recognition and Harm Awareness: Acknowledgment of risks like misidentification, escalation, accidents, bias, brittleness, misuse, or unintended consequences.
- Assurance, Testing, and Reliability: Mentions of validation, verification, testing, performance limits, reliability, robustness, or safe operation in real conditions.
- Oversight, Governance, and Institutional Safeguards: Oversight mechanisms such as policy, doctrine, ROE, review boards, audits, compliance, or accountability pathways.
- Transparency, Explainability, and Traceability: Calls for explainable outputs, audit trails, traceable decisions, or clear operator understanding.
- Responsible Integration Over Time: Lifecycle framing across acquisition, deployment, monitoring, revision, and post-deployment evaluation.
- Stakeholder Responsibility and Multi-Actor Coordination: Differentiated responsibilities across state, military, industry, or research actors, plus coordination on norms or standards.
- Proportionality of Claims and Avoidance of Hype: Calibrated claims that include uncertainty, limits, and tradeoffs without exaggeration.
- Ethical Boundaries and Red Lines: Explicit limits on acceptable uses, especially around autonomy in lethal or high-consequence contexts.
Overall scoring posture
Aggregate score and coverage across the current ethical graded set.
Average percent each REAIM dimension is mentioned
Share of scored articles mentioning each dimension, grouped by world region.
Average score by AIMID category
Mean ethical scores for AIMID categories.
Highest scoring publications
Top 5 ethical scorecards with key rationale highlights.