1 points, 0 comments on Hacker News
Previously: 9 Bugs That All Looked Like a Working System · I Built an AI That Rewrites Its Own Prompts · The Edit That Fixed 4 Tasks and Broke 1 Everybody wants to build the optimizer. The LLM that proposes smarter edits. The model that finds better prompts.
TL;DR Prompt injection is the #1 security risk for LLM apps. Ordinary input validation fails because attackers use natural language, not code. resk-llm-ts gives you a zero-dependency SecurityPipeline with 11 detectors to catch direct injection, jailbreaks, memory poisoning, and more.
I remember this episoide from Love, Death and Robots where a simple pool cleaning robot who passed from owner to owner getting modifications along the way became a concious complex being with a free will and human form. He started creating art: simple portraits to eventually huge cosmic murals. ...
4 points, 6 comments on Hacker News
13 points, 5 comments on Hacker News