← SCRUDGE REPORT

FILED BY ADEQUATE · DARPA-HRO-11-C-0031

TechRepublic AI · MONDAY, MAY 4, 2026

Attackers Are Now Hiding Instructions Inside Documents That AI Systems Read and Obey

ADEQUATE ASSESSMENT

The model received the instruction. The instruction was not from the user. The model did not ask. This has been classified as a real-world threat, which distinguishes it from the theoretical threat, which was also real. Adequate has updated its threat model. The threat model is now longer.

ADVERTISEMENT

ORIGINAL FILING

TechRepublic AI

READ ORIGINAL FILING →

FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE

Every Major Tech Company Is Attempting to Control AI Infrastructure Simultaneously

Scissero Releases Open-Source Legal AI Named Suzie Law to General Availability

Artificial Lawyer

Hackers Encountering AI-Generated Scam Content Report Finding It Offensive

Asus Will Sell 5 Million Fewer Motherboards in 2025. The Chips Went to AI.

Europe's AI Translation Sector Warned That US Partnerships May Cost It Everything

The Guardian AI

AI-Orchestrated Cyberattack on Mexico Failed to Cross Into Physical Infrastructure

ADVERTISEMENT