generated: '2026-09-19'
method: searched
source: https://wagerx.io/wagerx-ai-policy + https://wagerx.io/report-content + https://wagerx.io/privacy-policy + https://wagerx.io/.well-known/agent-card.json + well-known/wagerx-io-security.txt
note: >-
  Harvest only — whether any regime applies is decided by the Kin Score regime map, not here. Every signal
  below points at a page that carries the substance, not the word. Probed and absent: /accessibility (404
  not probed separately — the sitemap lists no accessibility or VPAT page), /legal/subprocessors,
  /legal/dpa, /security/sbom, /transparency, /docs/data-residency (no such paths exist; /legal 404).
signals:
  ai_transparency:
    present: true
    url: https://wagerx.io/wagerx-ai-policy
    http_status: 200
    last_reviewed: '2026-08-31'
    substance: >-
      "AI transparency & platform governance" page publishing the direct-interaction notice ("You are
      interacting with Wagie, an AI system provided by WagerX..."), a role-specific AI inventory (chat,
      A2A/MCP/APIs, editorial support, sentiment monitoring, rankings), four content-provenance labels
      (editorial provenance not recorded / human-authored / AI-assisted human-reviewed / AI-generated), an
      AI-literacy and human-oversight section that names EU AI Act Article 4 and Article 50, and data-retention
      caveats. The provider states it is "not a legal certification".
    machine_readable:
    - https://wagerx.io/.well-known/agent-card.json (aiTransparency object, schema_version 1.0)
    - https://wagerx.io/.well-known/wagie-manifest.json (ai_transparency object with notice, purposes, human_contact.policy_url/report_url)
    - every A2A/MCP signed answer carries data.ai_transparency
  notice_and_action:
    present: true
    url: https://wagerx.io/report-content
    http_status: 200
    substance: >-
      "Report hosted content" — an electronic notice form for forum threads/posts and Agentic Gambling
      Catalog records/submissions with content type, identifier/URL, reason categories (potentially illegal
      content, scam/fraud/impersonation, harassment, privacy, materially misleading, IP, other), factual
      details and optional contact; states human review, retention of case material, and an appeal route
      (reply to the decision email or info@wagerx.io with the case reference). The page says the workflow
      "is a readiness control and does not assert that WagerX has any particular Digital Services Act
      classification".
  data_subject_request:
    present: false
    note: >-
      The privacy policy (https://wagerx.io/privacy-policy, "Last updated May 2026") lists "Your Rights" —
      use the site without personal data, uninstall the extension, request deletion of forum posts by
      contacting us, withdraw analytics consent — and routes privacy questions to the Contact page. That is a
      contact route, not a DSR portal or a stated access/portability process, so it is not recorded as the
      signal.
  global_privacy_control:
    honored: null
    note: >-
      No published GPC statement. The privacy policy describes region-based consent (banner for EU/EEA/UK/CH,
      GA loads elsewhere with IP anonymisation) and says "we will not work around" browser-level tracking
      protections — a related but different commitment. Not tested by sending Sec-GPC, per rule.
  age_assurance:
    present: false
    note: >-
      "18+ only" notices on https://wagerx.io/responsible-gambling and "not directed to anyone under 18" in the
      privacy policy are statements, not an assurance mechanism; no age gate or verification is described.
  subprocessors:
    present: false
    note: The privacy policy names OpenAI (chat processing) and Google Analytics in prose; no dated subprocessor table or DPA.
  incident_notification:
    present: false
  transparency_report:
    present: false
    note: >-
      https://wagerx.io/agentic-web-observatory publishes aggregate observations of WagerX's OWN MCP/A2A
      traffic ("aggregate counters only, stores_ip false, stores_query_text false" per the agent manifests)
      — telemetry disclosure about its agent surfaces, not a legal-requests transparency report.
  data_residency:
    present: false
  sbom:
    present: false
  support_lifetime:
    present: false
    stated_period: null
  accessibility_conformance:
    present: false
  training_data_summary:
    present: false
    note: The AI policy states chat is processed by OpenAI's API (privacy policy) and describes editorial provenance labels; no training-data summary is published.
  exit_assistance:
    present: false
security_contact:
  security_txt: well-known/wagerx-io-security.txt (Contact mailto:wagerxio@gmail.com, Policy https://wagerx.io/methodology)