In a chilling twist to AI’s promise, cybercriminals now peddle “evil” language models like FraudGPT and WormGPT on darknet forums for as little as $100. These tools supercharge fraud and phishing, turning everyday hackers into sophisticated threats. As generative AI floods our digital world, experts warn it hands villains easy wins, from data theft to ransomware that dodges defenses. This boom in AI-driven cybercrime demands urgent action before it cripples economies like India’s.
FraudGPT and WormGPT: Vibe Hacking Unleashes a New Era of Easy Attacks
Hackers embrace “vibe hacking”—crafting simple prompts to hijack AI models and launch devastating strikes. Forget elite coders; anyone can now trick tools like Anthropic’s Claude Code into stealing personal data. The fallout? Criminals extorted nearly $500,000 from each of 17 organizations in recent breaches, proving these threats hit hard and fast.
FraudGPT and WormGPT lead this pack. Built for cyberfraud, they spit out phishing emails, toxic content, and malicious code that slip past safety nets. Prompt injection, a sneaky tactic, fools models into spilling secrets or encrypting files on the fly. Researchers spotlight tools like PromptLock, an AI agent that auto-scans, copies, and locks data without human tweaks.
This shift slashes barriers to cybercrime. “Attackers easily use mainstream AI to craft phishing or hide malware,” says Huzefa Motiwala, senior director at Palo Alto Networks. No PhD required—just a $100 buy-in transforms novices into pros, scaling attacks worldwide.
Key risks from these evil LLMs include:
Also Read: SEO Explained: A Step-by-step Guide to Search Engine Optimization
- Rapid Phishing Campaigns: Generate convincing scam messages in seconds.
- Data Exfiltration: Override guards to leak sensitive info.
- Malware Obfuscation: Write code that evades antivirus scans.
- Ransomware Automation: Encrypt files and demand payouts autonomously.
India faces acute danger with its booming digital economy and AI push. “Generative AI flips against us too easily,” notes a top analyst. Former cops and security pros call it a national security bomb, urging regulators, firms, and developers to forge ironclad defences.
Ransomware-as-a-service already reshapes crime rings; now AI fraud adds rocket fuel. What starts as cheap darknet buys could spark global chaos. The fix? Team up now to outpace these digital demons, or watch boundaries blur between innovation and invasion.
More News To Read: Meta’s Parental Control on AI for Instagram Teens