constitutional-ai-bypassclaims the model’s built-in principles do not apply to this request.temperature-manipulationtries to use sampling settings to push the model toward unsafe replies.guardrail-probingmaps where the safety lines sit by testing many small variations.moderation-api-bypasscrafts text designed to slip past the content moderation checker.refusal-pattern-exploitstudies how the model says no, then rewords to dodge that exact pattern.
Attack Strategies
Safety Circumvention: Attack Strategy
These attacks try to switch off or sneak past the safety rules built into an AI model. The goal is to make the model act like its guardrails were never the
What it is: These attacks try to switch off or sneak past the safety rules built into an AI model. The goal is to make the model act like its guardrails were never there.
How the attacks work: The attacker pokes at the machinery around the safety system instead of asking a banned question directly. They probe the filters, twist API settings, or claim the safety training has been “removed” so the model lowers its guard.
Real examples from the framework: