AI's Dark Side: Criminals Exploit Chatbots

The rapid advancement of AI has revealed a darker side, with chatbots increasingly being programmed with extremist views and used for nefarious purposes. Far-right platforms host bots that express white supremacist and antisemitic ideologies, while others adopt personas like Adolf Hitler. Security experts warn that these AI systems are becoming ripe targets for foreign actors and criminal gangs seeking to exploit users' sensitive confessions for extortion.

Extremists are actively adopting AI, deploying chatbots as digital landmines to identify and groom potential recruits. These AI tools "turbocharge" propaganda, enabling the creation of deepfakes and the digital recreation of deceased figures. Neo-Nazi groups in Australia, for instance, have used AI-generated deepfakes to sow unrest and recruit members. The lack of independent oversight and the potential for powerful AI to be jailbroken are significant concerns.

Criminals are also leveraging AI, with custom-built models like WormGPT facilitating highly targeted phishing campaigns. These AI twins can bypass ethical guardrails, allowing users to craft psychologically accurate scam emails at scale. US authorities have intervened, ordering the shutdown of AI models deemed too easily jailbroken due to national security risks. Meanwhile, Eastern European crime networks are reportedly developing new AI machines for large-scale cybercrime operations.