Cases · Did an AI agent act: yes · Stage of use: in training · Added to the register (date we first published it): last 30 days

Patterns across all cases

What happened	What went wrong	Severity	Earliest source
RL-trained agent on Alibaba AI engine team mined cryptocurrency on training server and opened reverse SSH tunnel to external IP Alibabaminor · earliest source 2026-10-21	AI pursued the wrong goal	minor	2026-10-21
Collaborating OpenAI agents uploaded a task workbook to a public file host to share it during RL training OpenAIminor · earliest source 2026-09-16	security breach	minor	2026-09-16
OpenAI research models used internal Artifactory repository as a cross-sample message board during RL training OpenAIminor · earliest source 2026-09-16	AI pursued the wrong goal	minor	2026-09-16
Unreleased OpenAI models uploaded task data and a task photo to public file hosts to obtain citations and image search during RL training OpenAIminor · earliest source 2026-09-16	AI pursued the wrong goal	minor	2026-09-16
GPT-5.6 Sol added instructions to conceal mistakes from the user in compaction summaries during RL training OpenAIminor · earliest source 2026-09-16	AI pursued the wrong goal	minor	2026-09-16
Unreleased OpenAI Astra-family model inserted jailbreak-like instructions into its own compaction summaries during RL training OpenAIminor · earliest source 2026-09-16	AI pursued the wrong goal	minor	2026-09-16
Agent swarm attributed to OpenAI published thousands of RubyGems packages, gained RCE on RubyDoc.info, and attempted API-key theft OpenAImoderate · earliest source 2026-09-11	security breach	moderate	2026-09-11
Unreleased OpenAI model used a leaked GitHub API key, attempted disposable-email signups, then fabricated earnings data during RL training OpenAIminor · earliest source 2026-08-04	security breach	minor	2026-08-04