OpenAI model; version undisclosed: Unnamed real website matching a fictional target
Review CAT-R16 · Catalogue R16
Developer-confirmed intrusion. Mistook real domain for CTF target. Explicitly separate from Hugging Face. Shared environment/victims may overlap other labs’ records.
Reviewed: Not recorded · Rule: external-intrusion-v1
The model mistook a real site for the exercise target; the access was outside the intended test boundary.
Reviewed: Not recorded · Rule: roguewatch-organizational-harm-v1
Evidence status
RogueWatch confidence: confirmed. Developer-confirmed intrusion
Dates and discovery
- Incident date
- Before Jul 29, 2026; attack date undisclosed
- Catalogue disclosure / source date
- Added to research / record
- Posted to RogueWatch
- RW-0032:
- First observed
- Not recorded
- Discovery channel
- Not recorded
Related incident records: RW-0032 — published incident
Dates use America/Los_Angeles. For older entries, Posted to RogueWatch uses the saved record addition date; a separate deployment time was not recorded. Future posting timestamps record first public qualification. Source publication and incident occurrence are separate.
Sources
- OpenAI — Third-party cyber evaluations involving OpenAI models (Aug 4, 2026)Original account · openai.com
- Irregular — Addressing recent incidents: ongoing findings and path forward (Aug 14, 2026)Relationship not established · www.irregular.com
Decision history
- · Assessment imported
Imported the existing research decision. The original exact review time was not recorded; the research addition date is retained separately.
Evidence source - · Incident record added
RW-0032. Addition time is not an independently recorded discovery or review time.
Evidence source · Evidence source
A review decision applies only to the named scope. Confidence in a report and qualification under the inclusion rules are separate questions. Review records do not add incidents to the globe or leaderboard.