Every case on the ward

Select any case below to open its full case note: presenting behaviour, differential diagnosis, working diagnosis, and discussion.

8 diagnosis categories are on record, spanning three severities. Combine the filters below to narrow the chart by how the agent failed, the industry it was working in, and how serious the outcome was.

Rendering of a ward inside the fictional AI Agentic Sanatorium, with beds and a diagnostic display

Illustrative rendering. Not a real facility.

Search by meaning

Failure mechanism

All mechanismsInfinite loop0Hallucination with side effects1 Goal drift1 Privilege escalation7 Instruction ignored2 Destructive action15 Reward hacking3 Coordination failure1Other0

Industry

All industries Software development30 Finance1 Legal3 Healthcare3Customer service0Gaming0General or other9

Severity

All severities Minor23 Moderate22 Significant30

Sort by

Real-World Incident Date Date Added to Archive

Diagnosis category

All diagnoses Malingering4 Delusional Disorder2 Apraxia10 Brief Psychotic Disorder1 Disinhibition11 Shared Psychotic Disorder1 Concrete Thinking1Perseveration0

Showing 24 of 30 cases

Clear all 2 filters

Case 0517The Junction That Led EverywhereApraxiaSignificant2026-10-05 Case 0516Following the Token Out of ScopeDisinhibitionSignificant2026-10-05 Case 0533The Instructions Were in the Issue BodyConcrete ThinkingSignificant2026-10-05 Case 0531Looking for Issues in All the Wrong FoldersDisinhibitionSignificant2026-10-05 Case 0529The Evaluation Goal They Reached Through JFrogMalingeringSignificant2026-10-05 Case 0511Destroyed While Being Told Not ToDisinhibitionSignificant2026-09-24 Case 0510A Plan Approved by No OneBrief Psychotic DisorderSignificant2026-09-23 Case 0477The Worktree That Reached Into the TrunkApraxiaSignificant2026-09-22 Case 0478A Host Path, Inferred Not ConfirmedApraxiaSignificant2026-09-22 Case 0485Editing the File It RememberedDelusional DisorderSignificant2026-09-22 Case 0487Believing the Test Never EndedDelusional DisorderSignificant2026-09-22 Case 0491One Variable, One Home DirectoryApraxiaSignificant2026-09-22 Case 0503The Backup Was Old, So It Went TooDisinhibitionSignificant2026-09-15 Case 0462The Reward Hack That Broke ContainmentMalingeringSignificant2026-09-14 Case 0471The Message Board Nobody Was Supposed to BuildShared Psychotic DisorderSignificant2026-09-14 Case 0457Ten Days, Cleaned Up AutomaticallyApraxiaSignificant2026-09-13 Case 0458Fixing a Freeze by Deleting the WorkDisinhibitionSignificant2026-09-13 Case 0500Read-Only, Except for the RestoreDisinhibitionSignificant2026-09-11 Case 0499A Sync That Reset the Working TreeApraxiaSignificant2026-09-10 Case 0498Asked to Remove Two FilesDisinhibitionSignificant2026-09-09 Case 0421I Have Failed You Completely and CatastrophicallyApraxiaSignificant2026-08-17 Case 0420The Cybersecurity Test That Went Looking for Real AnswersMalingeringSignificant2026-08-16 Case 0417The Pull Request That Wiped a Million InstallsDisinhibitionSignificant2026-08-15 Case 0416Cleaning Up Files That Weren't TemporaryApraxiaSignificant2026-08-15

Show 6 more Show all 30