Why text can look identical but behave differently
A hidden character can affect search, line breaks, sorting or display without looking like an ordinary letter. For example, alpha and al[U+200B]pha differ when the bracketed notation is replaced with an actual zero-width space. An editor with invisible-character display makes the difference easier to see.
Invisible does not mean malicious or AI-generated. Unicode includes characters needed for real writing systems and layout. The Unicode Standard defines these characters; a suspicious-looking character should be judged in context.
Check the exact position
Paste the text into the text cleaner. It lists supported findings by code point, line and column. Current checks cover U+200B zero-width space, U+2060 word joiner, U+00AD soft hyphen, several bidirectional controls, and U+FEFF when embedded after the start of the text.
Read the surrounding sentence before selecting a removal switch. A soft hyphen may be intentional in a publishing workflow; a bidirectional control may be needed to display mixed right-to-left and left-to-right text. If you are unsure, keep a copy and compare the rendered result after cleaning.
What this cannot tell you
Finding one of these code points does not show who wrote the text. Finding none does not prove there is no statistical watermark. Free AI Wiper checks supported characters, not hidden patterns spread across word choices or sentence structure. See the text format page for file and length limits.