A technique known as ASCII smuggling, initially developed to make AI prompt injection attacks more covert, is now being exploited by spammers to bypass email filters. This method uses invisible Unicode characters that are readable by computers but undetectable to humans. Microsoft Defender for Office 365 observed a dramatic surge in spam messages employing ASCII smuggling, with daily detections escalating from thousands to millions, indicating a significant shift in how malicious actors are attempting to deliver unwanted content. AI
IMPACT Spam campaigns are evolving to bypass AI-powered defenses, necessitating new detection strategies.
RANK_REASON A security technique developed for AI attacks is being repurposed for spam, with a significant increase in detection rates.
Read on Mastodon — mastodon.social →
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →