Fragmented enforcement and rapidly improving deepfake capabilities are increasing security risks to vulnerable groups and democratic resilience, while creating uneven legal standards that platforms can exploit through regulatory arbitrage. Deepfake regulation is likely to shift from ad hoc enforcement toward formal transparency, detection, and accountability requirements, with longer-term movement toward shared liability frameworks and election-security integration.