Analyzes images, videos, and audio files for signs of deepfake manipulation by checking metadata, provenance chains, and synthesis artifacts. Built by MEOK AI Labs with EU AI Act compliance features baked in, including risk management and bias detection capabilities. The server runs via stdio transport and installs through pip. You'd reach for this when you need to verify media authenticity in applications that handle user-generated content, journalistic verification workflows, or platforms that need to flag synthetic media under regulatory requirements. Part of a larger ecosystem of MCP servers focused on AI governance and safety monitoring.
claude mcp add --transport stdio csoai-org-deepfake-detector-mcp -- uvx deepfake-detector-mcp