Once popular for attacking AI, ASCII smuggling is embraced by spammers
Summary
A method called ASCII smuggling, originally used to hide harmful instructions in AI attacks, is now being used by spammers to bypass email filters. This technique hides text using special invisible Unicode characters that computers can read but people cannot see, allowing spam messages to avoid detection by security systems.Key Facts
- ASCII smuggling hides harmful or suspicious text using invisible Unicode tags that look like normal letters to computers.
- The method was first noticed two years ago to sneak malicious prompts past AI systems.
- Microsoft saw a huge rise in spam using ASCII smuggling, from about 21,000 cases per day in early February to over 2.5 million within a few days.
- Spammers insert invisible characters inside trigger words like “funding” to confuse email filters.
- This approach is more effective against machine learning and natural language processing spam filters.
- Traditional filters look for exact text strings, but ASCII smuggling changes the text-byte structure, making detection harder.
- Similar hiding tricks have been used before, like zero-width spaces, but ASCII smuggling uses a new range of Unicode tags.
- The sudden drop in ASCII smuggling spam was seen in mid-May 2026, after months of high activity.
Read the Full Article
This is a fact-based summary from The Actual News. Click below to read the complete story directly from the original source.
