# METR publishes its independent investigation (or an interim public report) of the Anthropic cybersecurity-evaluation incidents on or before 2027-01-11.

- **Artifact ID:** SO-2026-004
- **Status:** OPEN
- **Public record:** https://record.cybersecurityhq.com/forecasts/so-2026-004
- **Machine-record SHA-256:** `5dd3277bd5540cdae15492cca07d7268a9e5636f350bbc13bbc9e2c327a4f97d`

## Complete structured record

```json
{
  "id": "SO-2026-004",
  "class": "EVIDENCE_PUBLICATION",
  "claim": "METR publishes its independent investigation (or an interim public report) of the Anthropic cybersecurity-evaluation incidents on or before 2027-01-11.",
  "issued": "2026-10-13T12:30:00Z",
  "probability": 0.6,
  "benchmark": 0.5,
  "benchmark_kind": "ANALYST_PRIOR",
  "benchmark_basis": "Judgment. The engagement's initial term was eight weeks from early September (ends early November), extendable; METR publishes most engagements. Offsetting: extension by mutual agreement is explicitly provided for.",
  "adjustment": 0.1,
  "resolution_rule": "YES if metr.org publishes a report naming the Anthropic incidents, dated after issuance and on or before 2027-01-11.",
  "admissibility": "Only publications dated after 2026-10-13T12:30Z count; for EVENT-class, only events occurring after that timestamp count. Evidence pre-dating issuance resolves the forecast RETROSPECTIVE_DISCOVERY, not YES.",
  "exclusions": null,
  "resolves": "2027-01-11",
  "status": "OPEN",
  "outcome_date": null,
  "outcome_basis": null,
  "brier": null,
  "benchmark_brier": null,
  "resolution_source": "EXTERNAL",
  "experimental": true,
  "false_positive_cost": "Over-forecast would overstate third-party assurance timelines.",
  "ledger_hash": "a5d6703bed39e97a",
  "revision_log": [],
  "record_links": []
}
```