Taxonomy API

The taxonomy endpoint provides the complete domain and threat pattern hierarchy in a single JSON response.

Endpoint: /api/threats.json (opens in new tab)

Format: JSON (UTF-8)

Authentication: None required

CORS: Access-Control-Allow-Origin: *

Cache: public, max-age=3600 (1 hour)

Response Structure

The response contains a _meta object with versioning information, followed by a domains array. Each domain contains its nested sub_categories (threat patterns).

Top-Level Fields

Field	Type	Description
`_meta`	object	Version, generation timestamp, counts, licence
`domains`	array	8 threat domains, each with nested patterns

Domain Object

Field	Type	Description
`domain_code`	string	Stable identifier (e.g. `DOM-INF`)
`title`	string	Human-readable domain name
`slug`	string	URL slug (e.g. `information-integrity`)
`definition`	string	Full domain definition
`url`	string	Canonical domain page URL
`last_updated`	string	ISO date of last content update
`framework_mapping`	object	Cross-references to external frameworks (NIST, EU AI Act)
`sub_categories`	array	Nested threat patterns belonging to this domain

Pattern Object (within sub_categories)

Field	Type	Description
`pattern_code`	string	Stable identifier (e.g. `PAT-INF-001`)
`title`	string	Pattern name
`slug`	string	URL slug
`definition`	string	Pattern definition
`severity`	string	`critical` \| `high` \| `medium` \| `low`
`likelihood`	string	`increasing` \| `stable` \| `decreasing`
`url`	string	Canonical pattern page URL
`roles_affected`	array	Affected group slugs
`sectors_affected`	array	Sector slugs

Example Response

A truncated example showing one domain with one pattern:

{
  "_meta": {
    "name": "TopAIThreats.com Threat Taxonomy",
    "version": "1.0",
    "generated": "2026-02-26T00:00:00.000Z",
    "total_domains": 8,
    "total_sub_categories": 42,
    "url": "https://topaithreats.com",
    "license": "CC BY 4.0"
  },
  "domains": [
    {
      "domain_code": "DOM-INF",
      "title": "Information Integrity",
      "slug": "information-integrity",
      "definition": "Threats that compromise the reliability...",
      "url": "https://topaithreats.com/domains/information-integrity/",
      "last_updated": "2026-02-21",
      "sub_categories": [
        {
          "pattern_code": "PAT-INF-001",
          "title": "AI-Generated Disinformation Campaigns",
          "slug": "ai-generated-disinformation-campaigns",
          "definition": "Coordinated use of generative AI...",
          "severity": "critical",
          "likelihood": "increasing",
          "url": "https://topaithreats.com/patterns/ai-generated-disinformation-campaigns/",
          "roles_affected": ["consumers", "public-servants"],
          "sectors_affected": ["government", "media"]
        }
      ]
    }
  ]
}

Use Cases

Threat landscape dashboards — Consume the full hierarchy to build custom visualisations of AI threat domains and their patterns
Security tool integration — Map internal risk categories to the TopAIThreats taxonomy using domain codes and pattern codes
Research datasets — Extract severity and likelihood ratings across all 42 patterns for comparative analysis
Framework alignment — Use the framework_mapping field to cross-reference patterns with NIST AI RMF, EU AI Act, and ISO/IEC 42001

View live endpoint (opens in new tab) · All endpoints · View taxonomy