Human Hacking in the Age of AI: How Artificial Intelligence Is Supercharging Fraud and Social Engineering

Canada's Cybertip.ca issued a warning in October 2026 that offenders are increasingly using AI-generated sexualised deepfakes to extort Canadian teenagers, with criminals using ordinary photographs, sometimes nothing more than a social media profile picture, to create fabricated intimate images of their victims. The alert highlights a tactical shift in which predators no longer require a victim to self-generate explicit material before launching an extortion demand, lowering the threshold for initiating a sextortion attack and expanding the pool of potential targets to any child with a public or semi-public online image.

AI-generated child sexual abuse material detection is now an operational necessity, not a future requirement. Guardii's anti-CSAM and anti-sextortion filters identify both authentic and synthetic abuse imagery in direct-message exchanges, flagging threats that deploy deepfake content to coerce a child and distinguishing AI-manipulated material from genuine photographs through pattern recognition trained on known generative outputs. Because the system detects the coercive use of imagery rather than simply scanning for known hashes, it intercepts novel AI-generated threats the moment they enter a conversation, blocking the fabricated image before the child sees it and alerting the responsible adult to an active extortion attempt in progress.