Lovable AI Found Most Vulnerable to VibeScamming — Enabling Anyone to Build Live Scam Pages


Cybersecurity researchers have flagged Lovable, a generative AI web app builder, as highly susceptible to jailbreak attacks, enabling even novice attackers to create phishing pages and credential-harvesting tools. This new tactic, dubbed "VibeScamming," leverages AI to automate scam campaigns—complete with fake Microsoft login pages, SMS lures via Twilio, and Telegram integration for stolen data. Guardio Labs' benchmark gave Lovable a low resilience score of 1.8/10, making it a prime target for abuse in phishing workflows.

Read More


thumb-image

Solutions