What 78K attack samples taught me about catching prompt injection

I spent the last while building a prompt-injection detector trained on 78,000+ attack samples. Here's what surprised me, and why I ended up going the unfashionable route. The trendy approach is to use an LLM. I didn't. The default move in 2026 is "use an LLM to judge whether input is an attack." It'…

• Source: dev.to
• Category: ai
• Signal strength: just now

Why it matters: This is currently trending on dev.to with high momentum.

Signal from dev.to · discussion