Models Apply for AI Scam Video Gigs

Post LinkedIn

🔗Read original on Wired AI

#deepfakes #scams #ethicsai-face-models

💡AI deepfakes fuel 100 daily scam calls—arm your apps against this now

⚡ 30-Second TL;DR

What Changed

Dozens of Telegram channels advertise 'AI face models' jobs.

Why It Matters

This reveals rapid proliferation of AI deepfake scams via Telegram, raising ethical concerns for AI video tech. Practitioners must enhance detection to mitigate fraud risks in video applications.

What To Do Next

Integrate open-source deepfake detectors like DeepFaceLab into your video pipelines.

Who should care:Developers & AI Engineers

🧠 Deep Insight

Web-grounded analysis with 6 cited sources.

🔑 Enhanced Key Takeaways

•AI-enabled scams linked to Telegram-based Chinese vendors extracted 4.5 times more revenue ($3.2 million per operation) than non-AI scams, due to higher efficiency and scale[3].
•Deepfake-related scams caused $1.1 billion in global losses in 2025, tripling from 2024, with over 80% occurring on social media platforms like Telegram[6].
•Services now offer both real human 'substitute video models' and AI faces for pig butchering scams, targeting solo operators via monthly subscriptions on Telegram[1].
•Romance deepfake scams increasingly use real-time AI video calls and celebrity impersonations, such as fake Brad Pitt videos defrauding victims of over €800,000[6].

🔮 Future ImplicationsAI analysis grounded in cited sources

AI face model gigs will expand to include voice cloning services

Scammers already integrate AI voice cloning requiring only 3 seconds of audio with video deepfakes, enabling more scalable romance fraud operations[4].

Regulatory crackdowns on scam compounds will boost solo operator models

As Cambodia fragments large scam operations, Telegram channels promote independent 'franchise' rentals of human and AI models for bedroom-based fraud[1].

Deepfake scam losses will exceed $3 billion annually by 2027

2025 losses tripled from 2024 to $1.1 billion, driven by AI vendors' tools that increase per-scam revenue 4.5-fold and enable rapid scaling[3][6].