Home / Critical Brief / Category archive
Lemma Critical Brief · Category archive

AI Decision Integrity

Hallucination litigation, operational harm from misjudgment, ungrounded outputs.

4 Briefs
No. 011 · 2026-05-31

SynthID Watermark Reverse-Engineering

How a Statistical Attack Strips the Provenance Mark from AI-Generated Content

Pillar 01 Verifiable Origin Data Provenance AI Decision Integrity Brief →
No. 017 · 2026-05-31

McKinsey Lilli's Writable System Prompts

The Layer Governing the AI's Behavior Had No Integrity or Provenance

Pillar 02 Verifiable AI AI Decision Integrity Identity & AuthAgent Runaway Brief →
No. 012 · 2026-05-31

The Robert Williams Wrongful Arrest

When an AI Face-Match Drove a Government Enforcement Action Without Independent Verification

Pillar 02 Verifiable AI AI Decision Integrity Identity & AuthAI Bias / Harm Brief →
No. 005 · 2026-05-30

Noroboto Attack

AI Document Review Input-Integrity Forgery via Embedded Lying Fonts

Pillar 02 Verifiable AI AI Decision Integrity Data Provenance Brief →