AI Prompt & Inference Security

Secure your AI model from Adversarial Prompting and other attacks as well as scan inferences and other texts for potential biases, leaks, etc. Powered by Empsing Security Watchdog.

搜集汇总

背景与挑战

背景概述

该数据集名为'AI Prompt & Inference Security'，专注于保护AI模型免受对抗性提示等攻击，并通过扫描推理和文本来检测潜在偏见、泄露等安全问题。其技术由Empsing Security Watchdog提供支持。