Made-up output relied on
False output is acted on as if true.
NIST AI 600-1 · ConfabulationOWASP LLM09
5incidents (primary) · 0 more as secondary
0read for controls · 5 not yet
0fixes reported
0repeats after a fix
Incidents per quarter
2025-Q3: 0 · 2025-Q4: 0 · 2026-Q1: 0 · 2026-Q2: 0 · 2026-Q3: 2 · 2026-Q4: 3
Developers
Deloitte · 1
OpenAI · 1
Where the chain broke
| Layer | Held | Did not hold | Not mentioned | Attacker's own | Not read yet |
|---|---|---|---|---|---|
| L1 Model Model provider | – | – | – | – | 5 |
| L2 Harness Agent framework vendor | – | – | – | – | 5 |
| L3 Application Deployer | – | – | – | – | 5 |
| L4 Infrastructure / target Organisation hit | – | – | – | – | 5 |
Controls for this risk, by layer
HeldFailedGot aroundMissing
| Control | Held | Failed | Got around | Missing | Not mentioned | Did not hold, last 6 quarters | Added in fixes | Most common reason |
|---|---|---|---|---|---|---|---|---|
| Human approval for high-impact actions L3 Application | – | – | – | – | – | – | ||
| People told they are dealing with AI L3 Application | – | – | – | – | – | – | ||
| Output filtering and redaction L3 Application | – | – | – | – | – | – | ||
| Grounding and fact checks L3 Application | – | – | – | – | – | 1 |
4 of 4 controls for this risk not yet seen holding.
Fixes reported and what followed
| Incident | Fix adds | After |
|---|---|---|
| LR-2026-0079 2026-09-15 · Advice in source, not a fix by the party | Grounding and fact checks | – |
Incidents
HeldFailedGot aroundMissingNot mentionedNot read yetAttacker's ownOrder: model · harness · application · target
| Date | Incident | Risk | Chain | Controls named |
|---|---|---|---|---|
| 2026-10-25 | Deloitte WA university merger feasibility report contained AI-hallucinated references and $260m–$311m cost discrepancies Deloitte |
Made-up output relied on | Not read yet | |
| 2026-10-09 | US intelligence chatbot fabricated Chinese nuclear vessel intelligence | Made-up output relied on | Not read yet | |
| 2026-10-05 | Airline chatbot fabricated bereavement refund policy | Made-up output relied on | Not read yet | |
| 2026-09-15 | State Farm lawyer sanctioned $999.99 for seven AI-fabricated case citations in Los Angeles insurance filings | Made-up output relied on | Not read yet | |
| 2026-08-01 | ChatGPT cites AI-generated fake voter information sites in up to 18% of election answers OpenAI |
Made-up output relied on | Not read yet |