1
SafeGPT: Preventing Data Leakage and Unethical Outputs in Enterprise LLM Use
企业级LLM安全防护方案，双向过滤敏感数据与违规输出，为AI应用筑牢隐私与伦理防火墙。
arXiv:2601.06366v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are transforming enterprise workflows but introduce security an…