About AI Sec
Offensive AI security — prompt injection, jailbreaks, agent exploitation, red team writeups.
What this site covers
AI Sec is the index site for offensive AI security across this network. It collects the attack classes that matter against deployed models and agents, links each one to the disclosure or paper it came from, and points to the deeper write-ups on the specialist sites.
- Prompt injection, direct and indirect
- Model jailbreaks and guardrail bypasses
- Insecure output handling and the downstream sinks it reaches
- Agent and tool-use exploitation
- Model extraction and membership inference
- Red-team methodology drawn from published engagements
35 articles are published so far, indexed by attack class on the AI red teaming hub. New ones are announced on the RSS feed.
Free tools
Three interactive tools sit alongside the writing. They are static, run entirely in the browser, and require no account: the Attack Technique Atlas, a deep-linkable graph of attack families with prerequisites, incidents and defenses per node; the prompt injection scanner, which shows which detection patterns a given payload trips; and the AI Red Team Gym, a safe target for practising the payload-crafting loop. The glossary defines the terminology used across all of them.
How these articles are produced
Articles here are researched from primary sources: vendor and project documentation, published standards and specifications, research papers and preprints, and measurements published by whoever took them. Drafts are produced with AI assistance and then edited against those cited sources before anything is published.
No article on this site is based on first-hand testing in a private lab, and nothing here should be read as a measurement report of its own. Where a number appears, it comes from a source that is named, so you can check the original instead of taking this site's word for it.
Everything is published under a single editorial byline. That byline is a publishing identity for the site, not a claim about a named individual, and it does not carry professional credentials.
Corrections
Corrections are welcome. If something here is wrong, out of date, or attributed to the wrong source, email hello@aisec.blog with the page address and what it should say. Substantive corrections are made in the article itself rather than quietly dropped.
How this site is funded
This site currently runs no affiliate links, sponsored posts, display advertising or paid placements, and it sells nothing. The tools here describe higher-limit tiers as planned, but no page here shows a price or offers a checkout. If that changes, the disclosure page will say so.
Contact
Email: hello@aisec.blog
Site: aisec.blog
Published by: AI Sec Editorial
See also the privacy policy, the terms of use, and the editorial disclosure.