Invisible Unicode Technique Helps Phishing Campaigns Evade AI Email Filters
Sign in✓ Signed in
Microsoft has identified a high-volume phishing campaign that inserts invisible Unicode tag characters into financial lure words, helping messages evade some email-filtering and machine-learning detection paths while remaining readable to recipients. The technique, known as ASCII smuggling, was originally associated with hiding prompt-injection instructions from people while leaving them visible to language models. Microsoft documented the crossover in a Security Research report.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...