You’ve seen them popping up on blogs and portfolios lately. Those bizarre, glitchy, malformed fonts that look like a ransom note filtered through a broken CRT monitor. The creators call them “anti-AI fonts,” and they’re designed to stop scrapers from ingesting precious words into large language models.
It feels like a rebellion. A digital middle finger to the Silicon Valley machine. But here’s the brutal reality: You aren’t building a fortress against AI; you’re just locking your human readers outside in the cold.
Let’s be clear about what you’re actually doing when you deploy these typographical nightmares. You are actively sabotaging your audience to wage a futile war against a bot that doesn’t care.
Here’s the hilarious irony that most creators miss: by obfuscating your text to hide it from AI, you’re actually signaling that this content is highly valuable. You’re putting a neon sign on your work that screams, “This is worth stealing!” And the AI models? They don’t just rely on basic OCR. They use context, semantic understanding, and surrounding metadata to piece together exactly what you’ve written. You aren’t confusing the machine; you’re just making your site look like a broken Geocities page.
Security theater doesn’t stop algorithms. It just sacrifices your audience on the altar of intellectual property paranoia.
One commenter perfectly captured the absurdity of this trend, noting that we’ve entered a weird middle ground where we want to write things OCR algorithms have trouble transcribing—which also means we write things people with accessibility issues have trouble seeing. It’s the ultimate regression of the open web.
When you use a screen reader, those anti-AI fonts don’t translate into clever obfuscation. They translate into garbage. When you prioritize breaking an AI scraper over enabling a visually impaired reader, you’ve already lost the plot.
The cat-and-mouse game of spam, captchas, and AI content is exhausting. I get it. We all want to protect our work. But wasting your time on symbolic saber rattling—what IP trolls use to feel powerful—isn’t going to stop OpenAI or Google. It just tells your readers you care more about hypothetical scrapers than the real people looking at your site.
Write for humans. Let the machines read along. Anything else is just digital self-harm.
FAQ
Q: But doesn't making AI work harder waste their resources and cost them money?
A: It's a drop in the ocean. AI companies have billions in compute. You wasting their time by 0.01 seconds doesn't bankrupt them, it just makes your blog unreadable.
Q: What should I do instead to protect my content from AI?
A: Focus on value. If you're worried about AI taking your work, make your work more uniquely human—opinionated, insightful, and deeply connected to your community. Glitchy fonts don't build moats.
Q: Is the anti-AI font trend just a grift?
A: Largely, yes. It's symbolic saber rattling by IP trolls and security theater for paranoid creators. It solves an emotional need to 'do something' while practically achieving nothing.