Updates on Prompt Injection Attacks

Every AI story we track on Prompt Injection Attacks: 7 stories so far, each summarized in our own words and linked to the publisher that reported it.

Pulled from 125 sources

This month

Cybersecurity5 min read

AI Agents Can't Tell Instructions From Data—And Authentication Won't Fix That

Forbes14d ago

Cybersecurity5 min read

ChatGPT Flaw Let a Planted Prompt Send a Victim's Gmail Data to Another Account

The Hacker NewsSep 8

AI Research2 min read

Breaking Claude Code Opus 5 Auto Mode

Simon Willison's WeblogAug 27

Cybersecurity3 min read

Atlassian Rovo Exfiltrates Data, Bypassing Controls

Hacker NewsAug 5

Machine Learning3 min read

OpenAI built an AI super-hacker to break its own models, then locked it away

Covered by 2 sourcesJul 15

OpenAI unveils Lockdown Mode to protect sensitive data from prompt injection attacks

Covered by 3 sourcesJun 6

LLM & Text Generation6 min read

Improving instruction hierarchy in frontier LLMs

Open AI NewsMar 10

That's everything we have on Prompt Injection Attacks right now