Samsung said it is teaming up with deepfake-detection company Reality Defender to develop an on-device AI deepfake detection offering for enterprise users, running directly on Galaxy devices to help spot AI-generated fake voice, video and images and reduce impersonation scams and related risks. Reality Defender uses multiple AI models to analyze voice, video and image content in real time — for example flagging synthetic speech impersonating someone during a live call, or supporting customer identity checks and new-user onboarding. The project is set to launch first in Singapore, where the company cites surveys finding 56% of enterprises encountered audio deepfake fraud in 2024 and 52% faced video deepfake incidents; it also says victims of government-official impersonation scams in Singapore lost at least S$242.9 million in 2025, 26.6% of the year's total scam losses. Samsung says local processing reduces the need to send sensitive data to external servers, and plans to use its Knox security platform; no launch date or list of supported Galaxy models has been announced.