Over a Million Phishing Emails Used Hidden Text to Fool AI Security Filters
Researchers found that criminals are hiding innocent words inside malicious emails to confuse both traditional and AI-powered spam filters, and the technique is working.
Key points
- Barracuda Networks researchers observed more than one million retail-themed phishing emails using hidden text between April 2025 and the time of their report.
- The technique, called "text salting," hides harmless words inside an email's code so that security filters read clean content while victims see something else entirely.
- Criminals are using large language models, meaning AI tools that generate and edit text, to produce these deceptive emails faster than ever before.
- Current AI-based email security tools are not reliably detecting the technique, according to Barracuda vice president of information security Peterson Gutierrez.
More than one million phishing emails reached inboxes between April 2025 and the date of Barracuda Networks' findings. Every one used the same basic trick to get past security software. It's old. It's working.
The technique is called text salting. Criminals pepper the hidden code of a fake email with harmless words alongside the suspicious ones. Security software reads the code and sees "puppy" and "book" sitting next to "urgent" and "rewards," so it doesn't flag the message as dangerous. Victims see only what the criminal wants them to see.
Hiding that filler is straightforward. A word can be shrunk to a font size of zero, placed inside a box with zero dimensions, or pushed off the edge of the screen entirely. The reader never notices. The filter does, and that's the entire point.
Why aren't email security tools catching this?
Email security gateways, the automated systems that decide whether a message reaches your inbox, read underlying code rather than a rendered email. When criminals mix clean words into that code alongside suspicious ones, the filter gets a very different picture than the reader gets.
Security tools know about salting and try to strip hidden text before scanning. Criminals have responded by layering several hiding methods together, so that if one gets spotted, another still passes through.
AI-based filters are supposed to go further by grasping the meaning and tone of a message rather than just scanning for banned words. Gutierrez says that isn't happening. "Text salting appears simple, but its continued and growing use shows it remains effective," he told Dark Reading. "A technique many people associate with older spam-filter evasion can also influence modern AI-based detection." Worse, the same AI technology powering those defences is available to attackers: large language models let criminals write and modify deceptive text at a speed that wasn't possible a few years ago.
This asymmetry is worth watching. Our 8 July report on the EvilTokens ghost phishing campaign showed a different hiding strategy reaching inboxes by the same logic: keep the malicious payload invisible until it's too late for a filter to catch it.
The phishing emails themselves promise rewards and gift cards, pushing urgent deadlines to pressure people into clicking. The campaign, first reported by Dark Reading, impersonates retail brands, though Barracuda notes the fakes aren't especially convincing to a careful human eye.
Should you worry?
If you receive an unexpected email offering prizes or expiring rewards from a retailer, don't click any links. Go directly to the retailer's official website by typing the address yourself.
For security teams, Gutierrez puts it plainly: "Security tools need to evaluate the full context of the message, including the relationship between the visible content, hidden or excessive text, links, sender behavior, and the action the email is trying to prompt." Keyword scanning alone won't cut it anymore.
