Provably Safe AI (davidad agenda)

{
  "id": "provably-safe",
  "wikiId": "E484",
  "path": "/knowledge-base/responses/provably-safe/",
  "filePath": "knowledge-base/responses/provably-safe.mdx",
  "title": "Provably Safe AI (davidad agenda)",
  "quality": 65,
  "readerImportance": 49.5,
  "researchImportance": 29.5,
  "tacticalValue": null,
  "contentFormat": "article",
  "causalLevel": null,
  "lastUpdated": "2026-03-22",
  "dateCreated": "2026-02-15",
  "summary": "Davidad's provably safe AI agenda aims to create AI systems with mathematical safety guarantees through formal verification of world models and values, primarily funded by ARIA's £59M Safeguarded AI programme. The approach faces extreme technical challenges (world modeling, value specification) with uncertain tractability but would provide very high effectiveness if successful, addressing misalignment, deception, and power-seeking through proof-based constraints.",
  "description": "An ambitious research agenda to design AI systems with mathematical safety guarantees from the ground up, led by ARIA's £59M Safeguarded AI programme with the goal of creating superintelligent systems that are provably beneficial through formal verification of world models and value specifications.",
  "ratings": {
    "novelty": 5,
    "rigor": 5.5,
    "completeness": 7,
    "actionability": 4
  },
  "category": "responses",
  "subcategory": "alignment-theoretical",
  "clusters": [
    "ai-safety",
    "governance"
  ],
  "metrics": {
    "wordCount": 2248,
    "tableCount": 20,
    "diagramCount": 1,
    "internalLinks": 14,
    "externalLinks": 19,
    "footnoteCount": 0,
    "bulletRatio": 0.03,
    "sectionCount": 28,
    "hasOverview": true,
    "structuralScore": 15
  },
  "suggestedQuality": 100,
  "updateFrequency": 90,
  "evergreen": true,
  "wordCount": 2248,
  "unconvertedLinks": [
    {
      "text": "David \"davidad\" Dalrymple",
      "url": "https://davidad.org/davidad_cv.pdf",
      "resourceId": "d48f2cf5ef974f92",
      "resourceTitle": "[PDF] davidad_cv.pdf - davidad (David A. Dalrymple)"
    },
    {
      "text": "ARIA",
      "url": "https://www.aria.org.uk/programme-safeguarded-ai/",
      "resourceId": "kb-993eacf1d62c61ae",
      "resourceTitle": "ARIA Safeguarded AI Programme"
    },
    {
      "text": "<FBF entity=\"sid_XqjV4mbMXQ\" property=\"total-funding\">£59 million</FBF>",
      "url": "https://www.aria.org.uk/opportunity-spaces/mathematics-for-safe-ai/safeguarded-ai/funding/",
      "resourceId": "23035b1f8321f3c0",
      "resourceTitle": "Funding | Safeguarded AI"
    },
    {
      "text": "announced a significant pivot",
      "url": "https://www.aria.org.uk/insights/ai-progress-and-a-safeguarded-ai-pivot/",
      "resourceId": "kb-1ad0fbdc383198b6",
      "resourceTitle": "AI progress and a Safeguarded AI pivot"
    },
    {
      "text": "Nov 2025 pivot",
      "url": "https://www.aria.org.uk/insights/ai-progress-and-a-safeguarded-ai-pivot/",
      "resourceId": "kb-1ad0fbdc383198b6",
      "resourceTitle": "AI progress and a Safeguarded AI pivot"
    },
    {
      "text": "ARIA Safeguarded AI Programme",
      "url": "https://www.aria.org.uk/programme-safeguarded-ai/",
      "resourceId": "kb-993eacf1d62c61ae",
      "resourceTitle": "ARIA Safeguarded AI Programme"
    },
    {
      "text": "davidad's Programme Thesis",
      "url": "https://www.lesswrong.com/posts/qweXJ6v9heSn4wvdk/davidad-s-provably-safe-ai-architecture-aria-s-programme",
      "resourceId": "21e192e2e01e6f04",
      "resourceTitle": "Davidad's Provably Safe AI Architecture - ARIA's Programme Thesis"
    },
    {
      "text": "ARIA",
      "url": "https://www.aria.org.uk/",
      "resourceId": "kb-29e93dcb80065a4b",
      "resourceTitle": "ARIA – Advanced Research + Invention Agency"
    }
  ],
  "unconvertedLinkCount": 8,
  "convertedLinkCount": 3,
  "backlinkCount": 4,
  "hallucinationRisk": {
    "level": "medium",
    "score": 45,
    "factors": [
      "no-citations",
      "conceptual-content"
    ]
  },
  "entityType": "approach",
  "redundancy": {
    "maxSimilarity": 19,
    "similarPages": [
      {
        "id": "provable-safe",
        "title": "Provable / Guaranteed Safe AI",
        "path": "/knowledge-base/intelligence-paradigms/provable-safe/",
        "similarity": 19
      },
      {
        "id": "formal-verification",
        "title": "Formal Verification (AI Safety)",
        "path": "/knowledge-base/responses/formal-verification/",
        "similarity": 18
      },
      {
        "id": "ai-control",
        "title": "AI Control",
        "path": "/knowledge-base/responses/ai-control/",
        "similarity": 13
      },
      {
        "id": "probing",
        "title": "Probing / Linear Probes",
        "path": "/knowledge-base/responses/probing/",
        "similarity": 13
      },
      {
        "id": "safety-cases",
        "title": "AI Safety Cases",
        "path": "/knowledge-base/responses/safety-cases/",
        "similarity": 13
      }
    ]
  },
  "coverage": {
    "passing": 7,
    "total": 13,
    "targets": {
      "tables": 9,
      "diagrams": 1,
      "internalLinks": 18,
      "externalLinks": 11,
      "footnotes": 7,
      "references": 7
    },
    "actuals": {
      "tables": 20,
      "diagrams": 1,
      "internalLinks": 14,
      "externalLinks": 19,
      "footnotes": 0,
      "references": 5,
      "quotesWithQuotes": 0,
      "quotesTotal": 0,
      "accuracyChecked": 0,
      "accuracyTotal": 0
    },
    "items": {
      "summary": "green",
      "schedule": "green",
      "entity": "green",
      "editHistory": "red",
      "overview": "green",
      "tables": "green",
      "diagrams": "green",
      "internalLinks": "amber",
      "externalLinks": "green",
      "footnotes": "red",
      "references": "amber",
      "quotes": "red",
      "accuracy": "red"
    },
    "ratingsString": "N:5 R:5.5 A:4 C:7"
  },
  "readerRank": 307,
  "researchRank": 417,
  "recommendedScore": 159.94
}

No external links