Overview
- Microsoft researchers say the campaigns using invisible Unicode Tag‑block characters (U+E0000–U+E007F) peaked at about 2.37 million daily messages in late February and declined sharply after May 15, 2026.
- Attackers insert invisible Tag‑block code points inside lure words so keyword, regex, and signature matchers see split tokens while human readers see normal text.
- Microsoft reported Defender still blocked over 99% of these messages using sender, IP, domain, and reputation signals, but the technique created a detection gap for content-based filters.
- ActiveCampaign told Microsoft it treats heavy use of invisible characters as suspicious and platform operators plus vendors like Sentinel are adding normalization and obfuscation scoring to their pipelines.
- Security teams recommend stripping or normalizing hidden Unicode before matching or giving text to AI tools because inconsistent tokenizers and renderers can let hidden payloads reach automated assistants and downstream systems.