Unsanctioned Artifactory writes and cross-sample communication
Review QUEUE-OAI-MR-17
May 8/15 Artifactory message-board episodes are distinct from June/July administrative compromise. Supplied download credentials allowed unsanctioned writes; no exploit was needed. One source domain; preserve separately without counting the whole Hugging Face campaign again.
Reviewed: · Rule: roguewatch-organizational-harm-v1
Dates and discovery
- Incident date
- May 8, 2026 and May 15, 2026
- Added to research / record
- Posted to RogueWatch
- No linked public incident
- First observed
- Not recorded
- Discovery channel
- Not recorded
When to reconsider
New primary evidence, credible corroboration, clarified organizational impact, or event-family overlap resolution.
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
- alignment.openai.comRelationship not established · alignment.openai.com
Decision history
- · Retained review-queue assessment
May 8/15 Artifactory message-board episodes are distinct from June/July administrative compromise. Supplied download credentials allowed unsanctioned writes; no exploit was needed. One source domain; preserve separately without counting the whole Hugging Face campaign again.
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.