Deepfake harms, body-image risks & regulatory response

Synthetic-media harms continue expanding from sexual exploitation and school nudification into enterprise fraud, political manipulation, fabricated evidence, unauthorized likenesses, and targeted harassment. Reportedly weak DHS RIVR PAD performance and fragile watermarking reinforce the need for layered safeguards, provenance, consent, and human review rather than detector-only defenses.