Skip to content
EVIL ROBOTS
Menu
Books
Dossiers
Dispatches
Tech
Quiz
Glossary
Research
Bibliography
About
evilrobots.lol
/
Tags
/
Ai-Security
#
Ai-Security
Tagged · 1 entry
Research: Prompt Injection — The Unfixable Flaw
2026-03-15 · 5 min · research
The timeline of admissions that prompt injection cannot be solved. OpenAI, NCSC, Anthropic, Cisco, and the 'Attacker Moves Second' paper.