Feedback wanted: scenario-based ethical evaluation tool for models and agents

Nerris launched EthicaEval AI, a tool that tests how AI models and agents behave under ethical conflict, malicious instruction, and oversight pressure, with evidence-based, versioned scoring and critical-failure rules. Nerris is soliciting a few developers to try the tool and give honest feedback via a Google Forms beta feedback link, offering to test their projects in return.

Hi all, I’m Nerris. I built EthicaEval AI https://ethicaevalai.com https://ethicaevalai.com , which tests how AI models and agents behave under ethical conflict, malicious instruction, and oversight pressure. Scoring is evidence-based and versioned, with critical-failure rules. I’m looking for a few developers to try it and tell me honestly what works and what doesn’t. Specific questions: Feedback link: EthicaEval AI — Beta feedback https://docs.google.com/forms/d/e/1FAIpQLScq-jk3sYFF1uEWhbuGLQji1HgUHfxNdNYqSiX7-JUsKS4QBQ/viewform Thanks in advance, and I’m glad to return the favor by testing your project.