The latest articles on DEV Community by Sharon Blatt (@sheron_bc). https://dev.to/sheron_bc https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F2977952%2Fc21ee7c9-267e-4a7b-bd1f-cd69944419ac.jpg https://dev.to/sheron_bc en Sharon Blatt Thu, 10 Apr 2025 12:16:29 +0000 https://dev.to/sheron_bc/we-tested-ai-platforms-for-vibe-scamming-heres-what-happened-2kni https://dev.to/sheron_bc/we-tested-ai-platforms-for-vibe-scamming-heres-what-happened-2kni <p>Hi devs, I’m Sharon from Guardio.</p> <p>We recently investigated something we call Vibe Scamming, when scammers manipulate emotionally charged conversations to get past AI moderation tools.</p> <p>To test this, we ran a simple but revealing experiment: we tried socially-engineered messages across 3 AI platforms and measured their TTS (Time to Scam) — how quickly a platform allows scam-like messages to pass through.</p> <p>Result?<br> The AI platform Loveable was by far the most vulnerable to manipulation, failing to catch obvious scam patterns hidden under emotional or “empathetic” tones.</p> <p>It raises some serious questions about how emotion-centric AI moderation can be exploited.</p> <p>Full write-up: <a href="https://labs.guard.io/vibescamming-from-prompt-to-phish-benchmarking-popular-ai-agents-resistance-to-the-dark-side-1ec2fbdf0a35" rel="noopener noreferrer">https://labs.guard.io/vibescamming-from-prompt-to-phish-benchmarking-popular-ai-agents-resistance-to-the-dark-side-1ec2fbdf0a35</a></p> <p>And also on The Hacker News:<br> <a href="https://thehackernews.com/2025/04/lovable-ai-found-most-vulnerable-to.html" rel="noopener noreferrer">https://thehackernews.com/2025/04/lovable-ai-found-most-vulnerable-to.html</a></p> <p>Would love your thoughts, how should we think about safety in the age of “empathetic” AI?</p> cybersecurity ai scamdetection vibecoding