拉德韦推出LLM防火墙,保护企业AI免受快速注入和数据泄露. Radware launches LLM Firewall to protect enterprise AI from prompt injections and data leaks.

Radware推出了LLM防火墙,这是一个新的安全工具,旨在迅速阻止诸如及时注射和越狱等威胁,从而保护企业基因化人工智能系统。 Radware has launched LLM Firewall, a new security tool designed to protect enterprise generative AI systems by blocking threats like prompt injection and jailbreaks at the prompt level. 模型-不可知性解决方案融入其云应用保护服务,提供实时的、AI驱动的防御,以防止数据渗出,确保遵守GDPR和HIPAA等条例。 Integrated into its Cloud Application Protection Services, the model-agnostic solution provides real-time, AI-driven defense to prevent data exfiltration and ensure compliance with regulations such as GDPR and HIPAA. 它处理2025年OWASP前十大LLMs的风险,LLMs作为“LLMs的WAF”运作,以确保跨平台的AI工作流程,同时又不干扰创新。 It addresses risks outlined in the 2025 OWASP Top 10 for LLMs, functioning as a "WAF for LLMs" to secure AI workflows across platforms without disrupting innovation.