Anti AI

How Anti AI works

Anti AI checks posts, images and videos as they load and gives each a score for how likely it is to be AI-made.

• Platform labels such as "AI info" or "Altered or synthetic content" are the strongest signal.
• Image metadata from AI tools (content credentials, generator names).
• Known AI sites that serve or host generated media.
• Writing patterns common in AI text. This is a best guess and will catch some human posts.

Sensitivity sets how much evidence is needed. Paranoid flags almost anything suspicious; Relaxed only acts on near-certain cases.

Label marks the item, Blur hides it behind a Show button, Remove hides it completely.

Everything runs on your computer. Page text never leaves your browser.