Tweets Are Training AI Without Permission

Twitter/X posts are a main ingredient for training Large Language Models. Most users never gave permission, never knew their words were used, and cannot retract them. This creates an urgent problem of personal data decay that developers rarely consider.