Cloudflare's 'Identity-Aware AI Gateway' tracks requests sent by employees and AI agents to AI models in real-time, and provides identity-based access control, cost limit settings, and sensitive information auto-blocking capabilities on a single platform.

The Identity-Aware AI Gateway integrates with Cloudflare Access to link all AI requests to verified user identities and can be deployed immediately without building separate systems, the company explained.

Additionally, it identifies simple tasks being processed by high-performance models and suggests points for transition to low-cost models, while caching repeated requests reduces unnecessary costs. Sensitive information such as employee names and passwords is automatically removed before being transmitted to external AI providers.

Dane Knecht, Chief Technology Officer at Cloudflare, stated, "By directly connecting Cloudflare's access control capabilities with the AI Gateway, each team can freely use the desired AI models, while the IT team can secure real-time visibility, protective measures, and necessary budget control functions."

The core features provided by the Identity-Aware AI Gateway include △real-time identification of AI request originators and transmission contents △anomaly detection and alerts by user △cost reduction analysis through high-performance and low-cost model switching △identity authentication based on IdP and Zero Trust Network Access (ZTNA) integration △unified traffic management across multiple AI model providers.

Max Baumgarten, Security Engineer at Flexport, stated, "By integrating Cloudflare Access with the AI Gateway, all requests are assigned verified identities, and each team can safely adopt necessary AI services without building separate authentication systems for each AI tool."