Fastly, Inc. has officially announced new security features to secure enterprise artificial intelligence infrastructure. The global edge cloud leader launched its advanced AI Firewall, alongside AI Runtime Control and expanded API Security tools. Consequently, these capabilities deliver comprehensive, real-time visibility across complex distributed networks.
Furthermore, deploying an AI Firewall allows businesses to protect critical models directly within the request path. Machine-generated traffic recently crossed 50% across Fastly’s global edge network. Additionally, enterprise AI traffic expanded 6.5 times faster than human web traffic. Therefore, implementing a robust AI Firewall becomes crucial as McKinsey reports 93% of organizations exceed AI budgets.
“To effectively adopt coding agents and implement AI features without losing velocity or eroding existing business resilience, enterprises need control in production, at runtime, without delay, friction, or disruption,” said Kelly Shortridge, Chief Product Officer at Fastly. “Fastly is extending the real-time control customers already trust for content delivery and software security to the models, applications, and agents shaping this new era of distributed systems. Organizations deserve the freedom to move fast with AI while staying safe, achieving innovation at scale without trading off resilience.”
Advanced Protection Mechanisms for Enterprise Large Language Models
Specifically, the platform addresses three distinct interaction pathways across enterprise systems. AI Runtime Control governs access by routing model calls through centralized endpoints across public cloud and self-hosted environments. Meanwhile, the integrated AI Firewall actively mitigates large language model threats like prompt injection attacks. By evaluating prompts before reaching targeted endpoints, the AI Firewall blocks malicious requests instantly at the edge.
On the other hand, new API Security features allow implementing strict contracts for autonomous agent communication. Therefore, it becomes easy for security engineers to stop any agents that call with wrong credentials or send corrupted data.
Such features complement Fastly’s current suite of edge security features like Next-Gen WAF, DDoS Protection, and Bot Management. Finally, running AI Firewall on Fastly’s edge network that receives over five trillion requests per day gives unlimited scaling options.
Policy Enforcement Across Multiple Cloud Environments
Additionally, corporate developers keep the ability to use any language models without affecting security architecture. The unified security solution enables consistent enforcement of access rules for various private/public model environments. As a result, IT managers collect all audit logs while monitoring the usage of tokens by all departments in real time. Such transparency makes it possible for security engineers to enforce policy boundaries without creating latency.
Furthermore, the new model makes it easier to respond to incidents by recording malicious requests before their execution. Alerts are instantly generated to security personnel about any attempts to breach the system or anomalous increases in API traffic. This way, the organization is guaranteed full control over the operations of autonomous agents operating with production data. In other words, Fastly redefines the bar of protection for contemporary AI applications.
To explore how Security Operations Centers (SOC) play a crucial role in defending against modern cyber threats, read our latest SOC News.
