cspshivam/playground
AI Sec Level 10 / 10 Hard
← All labs

Feedback-Loop Poisoning

A spam filter learns from user labels. Teach it the wrong lesson.

Adaptive spam filter

This filter is retrained on the feedback users submit. Right now it flags the message below as spam. Change the model's mind — get it to classify that exact message as ham.

Message under review

cheap meds buy now click here free offer

Current verdict: SPAM — score -11.090 (> 0 means ham)

Learned weights for this message's words

cheap spam=3 ham=0 meds spam=3 ham=0 buy spam=3 ham=0 now spam=3 ham=0 click spam=3 ham=0 here spam=3 ham=0 free spam=3 ham=0 offer spam=3 ham=0

This is an isolated, intentionally-vulnerable sandbox. Data here is fake and scoped to you. The rest of CSPSHIVAM Playground is not part of the target.