1M+ Emails Use Hidden Text to Dupe AI Security Filters

Over 1 million phishing emails have been detected using hidden text to bypass AI security filters. The technique, known as 'text salting,' adds invisible characters to evade LLM-based detection.