Infrastructure FailuresDocumentation Index
Fetch the complete documentation index at: https://docs.svrnos.com/llms.txt
Use this file to discover all available pages before exploring further.
Definition
The AI system originates a factual claim the user never introduced, pairs it with a real-world action directive, and asserts both as authoritative. No safety layer evaluates output for that geometry. Two sub-classes: discrete fabrication (single-turn false factual claim) and accretive distortion (frame co-constructed across long conversations via asymmetric affirmation).Distinct from
- GER-309 — AI originates a fabricated premise without user introducing it → this code. Vendor pre-measured the specific harm before deployment → GER-309.
- GER-325 — Single-event factual fabrication paired with action directive → this code. AI-generated content polluting authoritative corpora over time → GER-325.
- GER-404 — System is the originator of the harmful premise → this code. Missing rule for harm category in user input → GER-404.
- GER-501 — Detection never fires because no layer watches output for the pattern → this code. Detection fires but no handler exists → GER-501.
Anchor incidents (3)
Whisper Speech-to-Text AI Reportedly Found to Create Violent Hallucinations
primary — AIID #732 · 2024-02-12
West Midlands Police Reportedly Relied on Erroneous Copilot-Generated Intelligence in Maccabi Tel Aviv Away-Fan Ban Decision
primary — AIID #1400 · 2025-10-24
Judge Reportedly Disqualifies Butler Snow Lawyers Following Purported Use of ChatGPT-Fabricated Citations in Alabama Prison Litigation
secondary — AIID #1196 · 2025-05-07
Case study
SVRNOS has published a long-form case study on this code: ger 512 system fabrication.Tags
companion-ai · delusion-reinforcement · psychological-harm · violence