'As adoption grows, confidence in safeguards must rise with it': Microsoft reveals new tool which can track backdoors in LLMs - and it's hoping this will restore trust in AI across the world Microsoft introduced a scanner that detects poisoned open-weight language models by analyzing attention behavior, memorization leaks, and trigger flexibility. #memorization_leaks #adoption_grows #attention_behavior #trigger_flexibility #Microsoft_reveals #world_Microsoft #Microsoft_introduced #track_backdoors #restore_trust #detects_poisoned
