SocietyAI glossary
Red teaming
Deliberately attacking your own system to find its flaws before others do: booby-trapped documents, borderline requests, exfiltration scenarios. You fix, and you start again.
In other words
Paying someone to burgle your own home, while you can still change the lock.
Alyriaz articles that mention it
- What is "red teaming"? Attacking your own model before others do.
Red teaming involves deliberately attacking your own system to uncover its vulnerabilities before they can be exploited. Here's how it applies to AI, what it uncovers, and why it isn't enough.