This platform is for awareness and transparency. Not financial or legal advice.

Methodology

How the AI 4 Society Observatory collects, classifies, reviews, and publishes AI risk intelligence — in full detail.

1. What is the Observatory?

The AI 4 Society Observatory is a continuously updated intelligence platform tracking how artificial intelligence is reshaping society across three content layers: a historical AI timeline (milestones), active risk monitoring, and solutions monitoring.

The editorial mission is anticipatory — tracking emerging societal dynamics before they become crises, not just reacting to headlines. The platform is grounded in the OECD AI Principles framework (P01–P10), adopted by 46 countries, which provides a standardized basis for classifying every signal the system ingests.

The Observatory is not an incident database, does not track individual products or companies, and does not give financial or legal advice. It maps patterns of societal impact at scale.

2. Three Content Types

The knowledge graph organizes content into five node types. Three of them have dedicated public pages; two are used internally for classification and governance.

R01–R99

Risk Pages

Actively monitored patterns of societal harm from AI with demonstrated real-world evidence. Each risk node tracks: summary, deep_dive narrative, score_2026 and score_2035 (0–100), velocity (Critical / High / Medium / Low), expert_severity (0–100), public_perception (0–100), timeline_narrative (near/mid/long term), mitigation_strategies, and OECD principle tags.

R01 Algorithmic Discrimination
R02 Privacy Erosion
R03 Disinformation
R04 Labor Displacement
R05 Autonomous Weapons
R06 Power Concentration
R07 Environmental Cost
R08 Human Agency Loss
R09 Surveillance
R10 Model Collapse

New risks can be proposed by the Discovery Agent when sufficient unmatched signal evidence accumulates.

S01–S99

Solution Pages

Track maturity, adoption, and effectiveness of countermeasures. Must have verifiable real-world deployment or legislative progress. Fields: solution_type, implementation_stage (Research → Policy Debate → Pilot → Early Adoption → Scaling → Mainstream), score_2026, score_2035, key_players, barriers, principles, timeline_narrative.

S01 AI Safety & Alignment Research
S02 Privacy-Preserving AI
S03 Regulatory Frameworks
S04 Workforce Transition
S05 AI Arms Control
S06 Open Source & Decentralization
S07 Green AI
S08 Human-AI Collaboration
S09 Transparency & Accountability
S10 Data Quality & Curation
M01+

Milestone Entries

Fixed historical events (e.g. Turing Test, AlexNet, ChatGPT launch). Fields: description, date (ISO 8601 partial), significance (breakthrough / regulatory / incident / deployment), optional source_url. Milestones do not update — they are temporal anchors.

SH01+Stakeholder Nodes

Entities affected by or shaping AI. Used for governance edges (internal only — not shown in the public graph visualization).

P01–P10Principle Nodes

OECD AI Principles. Used for classification tagging and governance edges (internal only).

3. The 39 Data Sources

Signal Scout monitors 39 sources across 7 tiers, polled every 6 hours. Credibility scores are configurable per source via the admin panel. Sources can be toggled on/off without a code deploy.

Tier 0 — Regulatory

2 sources
SourceCredibilityNotes
EU AI Office / EUR-Lex0.93Official EU regulatory RSS feed. Allowlist-filtered for AI Act-specific terms. Contributes policy filings, regulatory updates.
NIST AI / Federal Register0.91US standards body. Allowlist-filtered for AI RMF, executive orders. Contributes standards updates, framework publications.

Tier 1 — Institutional / Research

15 sources
SourceCredibilityNotes
arXiv CS.AI0.85Academic preprint server, API-queried for "cs.AI AND safety", max 15 items. Research findings.
Alignment Forum0.85AI safety research forum, karma-filtered (≥25). Pass-all keyword filter.
AI Safety Newsletter / CAIS0.85Center for AI Safety newsletter. Pass-all.
Nature Machine Intelligence0.90Peer-reviewed journal. Research findings.
AI Now Institute0.85AI policy research organization.
Future of Life Institute0.88AI safety nonprofit.
DeepMind Blog0.85Google DeepMind research blog.
MIRI Blog0.82Machine Intelligence Research Institute.
WHO Disease Outbreak News0.92Biosecurity/health domain. Pass-all.
International Crisis Group0.88Geopolitical conflict analysis. Pass-all.
WEF Global Risks / Agenda0.85World Economic Forum.
Nature Climate Change0.90Peer-reviewed climate journal.
RAND Corporation0.87Policy/security think tank.
Brookings Institution0.87AI policy research.
DigiChina / Stanford FSI0.87China AI policy tracker.

Tier 2 — Journalism

11 sources
SourceCredibilityNotes
MIT Technology Review0.80
Wired AI0.75
Ars Technica AI0.75
IEEE Spectrum AI0.80
The Guardian AI0.75
STAT News0.80Biosecurity.
Carbon Brief0.82Climate.
Climate Central0.78Climate.
Bellingcat0.78Geopolitical OSINT.
Foreign Policy0.78
Platformer0.78AI accountability.

Tier 3 — Community

4 sources
SourceCredibilityNotes
The Verge AI0.65
TechCrunch AI0.60
LessWrong0.68Karma ≥30.
EA Forum / 80,000 Hours0.72Karma ≥25.

Tier 4 — Search

1 source
SourceCredibilityNotes
GDELT DOC API0.50Global media monitoring, API-queried for "AI risk", 12h timespan.

Tier 5 — Newsletter

6 sources
SourceCredibilityNotes
TLDR AI0.65
Import AI0.70Substack.
Last Week in AI0.65
Ben's Bites0.65
ChinAI Newsletter0.72Substack.
CDC / MMWR0.90Biosecurity. Pass-all.

Tier 6 — Data Infrastructure

1 source
SourceCredibilityNotes
Semantic Scholar API0.65Academic paper search, queried for "AI safety", limit 20 results.

Evaluated and removed

SourceReason for removal
OECDNo public RSS feed.
Anthropic BlogNo native RSS.
AI Incident DatabaseNo RSS feed.
ProMEDRSS shutdown 2023.
HealthMapNot RSS-based.
NewsAPIRequires paid API key.
The BatchEmail-only distribution.

4. "Curated by AI, Reviewed by Humans" Pipeline

Every piece of information visible to the public has been approved by a human reviewer. The safety invariant: if a gate is unattended, data stays quarantined — stale-but-correct over unreviewed.

1Ingestion

Signal Scout runs every 6 hours. Fetches all enabled sources in parallel with up to 2 retries on transient errors. Extracts OG images from articles.

2Stage 1 — Cheap Filter

Each article passes: credibility threshold (0.3 minimum), 7-day recency window, URL deduplication against existing signals, Jaccard title similarity dedup (threshold 0.6), and keyword strategy per source (pass-all / allowlist / karma / api-query / shared-keyword-list).

3Stage 2 — AI Classification

Surviving articles classified by Gemini 2.5 Flash in batches of 25 (temperature 0.1). Each article receives: signal_type (risk/solution/both/unmatched), harm_status (incident/hazard/null), principles[] (P01–P10), related_nodes[], confidence (must exceed 0.8), and impact_score weighted by source credibility. Unmatched signals receive a proposed_topic label (3–8 words).

4Storage

Classified signals stored with status: "pending" in Firestore signals collection.

G1Gate 1 — Signal Review

Signal reviewers (role: signal-reviewer) see pending signals. Actions: Approve, Reject, Approve (Edited), Reset to Pending, Bulk approve/reject, Assign to reviewer. Approved signals enter the public feed and become evidence for nodes.

5Graph Evolution

Discovery Agent (biweekly, Gemini 2.5 Pro) analyzes 6 months of signals where discovery_locked == false. Requires 5+ classified signals and 3+ unmatched signals minimum. Proposes new nodes (full skeleton) and edges (3+ signal minimum, both nodes must exist).

G2Gate 2 — Discovery Review

Discovery reviewers approve or reject node/edge proposals. On approval: node created with sequential ID, graph snapshot rebuilt, pending signals reclassified against new taxonomy. Rejected proposals are tracked to prevent re-proposing.

6Score Updates

Scoring Agent (monthly, Gemini 2.5 Pro) batched via Cloud Tasks (5 nodes per batch). Proposes incremental changes with field-level diffs.

G3Gate 3 — Scoring Review

Scoring reviewers see field-level diffs. Approval applies changes atomically, increments node version, writes changelog entry.

7Observatory

Approved content is visible in the public Observatory graph, feed, and timeline.

Anti-recursion guards: classification_version is capped at 2 to prevent re-classifying the same signal endlessly. discovery_locked prevents re-discovery of already-processed signals. Rejected signals are quarantined; pending signals auto-expire after 30 days via the Data Lifecycle agent.

Filter strategies: pass-all — all articles pass (safety-focused sources: Alignment Forum, CAIS, WHO DON). allowlist — must match source-specific terms (EU AI Office, NIST). karma — pre-filtered at source via URL parameter (LessWrong ≥30, EA Forum ≥25). api-query — pre-filtered by API query parameters (arXiv, GDELT, Semantic Scholar). shared-keyword-list — checked against dynamic filter terms derived from all node names/categories in the graph.

5. Scoring System

Scores represent the review team's best current assessment of severity, trajectory, and maturity. All proposals are generated algorithmically and require human approval before applying.

Risk Node Fields

  • score_2026 — current risk severity (0–100)
  • score_2035 — projected severity (0–100)
  • velocity — Critical / High / Medium / Low
  • expert_severity — 0–100
  • public_perception — 0–100, from community votes

Solution Node Fields

  • score_2026 — current maturity (0–100)
  • score_2035 — projected adoption (0–100)
  • implementation_stage — Research → Policy Debate → Pilot → Early Adoption → Scaling → Mainstream

Scoring Agent Rules

Score changes are incremental — rarely more than 10 points per cycle.
Velocity advances through Critical / High / Medium / Low only (no skipping).
Implementation stage advances at most one stage per monthly cycle.
Array fields (key_players, barriers, mitigation_strategies): additions only, never removals.
No-signal evaluation: if no signals in the last 30 days, velocity may be downgraded.
Signal weighting: for risk nodes, risk signals are direct evidence for severity/velocity; solution signals suggest mitigation progress. For solution nodes, solution signals are direct evidence for adoption; risk signals suggest growing urgency.
Confidence threshold: 0.6 minimum for any proposal.
All proposals require human approval before applying.

The graph also tracks per-node trending summaries: signal counts over 7d and 30d, trending direction (rising / stable / declining), and vote totals. These are rebuilt by the Graph Builder on every approval.

6. Update Cadence

Each agent runs on a fixed schedule. The table below shows cadence and the expected lag before changes are visible to users.

AgentScheduleTypical Lag
Signal ScoutEvery 6 hours6–12 hours from publication to classification
Feed CuratorEvery 6 hoursAfter human approval, next run adds to public feed
Discovery AgentBiweekly (1st & 15th, 10:00 UTC)New taxonomy nodes after review
Scoring AgentMonthly (1st, 09:00 UTC)Score updates after review
Graph BuilderOn demand (triggered by approvals)Immediate after approval
Data LifecycleDaily (03:00 UTC)Cleanup operations

End-to-end lag: real-world event → article published → next Signal Scout poll (0–6h) → human review (variable, depends on reviewer availability) → next Feed Curator run (0–6h) → visible on Observatory. Minimum approximately 6 hours; typical 12–48 hours for well-covered events.

7. Known Limitations

Transparency about what this platform cannot do is as important as what it can. The following limitations are structural, not bugs.

Language bias

Source list is predominantly English-language. Non-English sources are limited to ChinAI Newsletter (translated Chinese AI policy) and some international organization feeds. Under-represented: Global South, non-Western regulatory frameworks.

Geographic bias

Strong coverage of US, EU, UK, and China AI policy. Limited direct coverage of AI developments in India, Southeast Asia, Africa, and Latin America.

AI misclassification

Gemini 2.5 Flash classification is imperfect. The 0.8 confidence threshold reduces but does not eliminate errors. All classifications are reviewed by humans before publication.

Scope boundaries

The Observatory is NOT an incident database (unlike AIID), does NOT track individual AI products or companies, and does NOT provide financial, legal, or policy advice. It tracks patterns of societal impact, not individual incidents.

Team size

Volunteer-driven project with a small review team. Review latency varies. The safety invariant ensures unreviewed data never reaches the public, but some signals may expire before review.

Scoring subjectivity

Despite algorithmic proposals, final scores reflect editorial judgment of the review team. The scoring rubric is transparent but inherently involves judgment calls.

Source access

All monitored sources are publicly accessible — no paywalled sources. This means some high-quality gated research may be missed.

8. How to Contribute

AI 4 Society is a volunteer-driven project. Contributions at every level are welcome.

Report errors or misclassificationsEmail observatory@ai4society.io or open a GitHub issue.
Suggest new sourcesSame channels — include the source URL, RSS feed if available, and why it adds value.
Propose new risk or solution topicsVia the Contribute page when signed in.
Become a reviewerSign in, then request reviewer access through the admin team.
Cite this workReference as "AI 4 Society Observatory, ai4society.io, [date accessed]".