Global edge cloud platform leader Fastly, Inc. announced the launch of AI Runtime Control, AI Firewall, and enhanced API Security capabilities. Engineered to deliver real-time visibility, security, and governance across enterprise artificial intelligence deployments, the unified suite allows organizations to safely manage model routing, control financial spend, and enforce strict security boundaries without sacrificing operational velocity.
The release arrives as enterprise AI transitions rapidly from early-stage experimentation to production-grade deployments. On Fastly’s global network, machine-generated traffic surpassed 50% in July and August 2026, while AI traffic grew 6.5 times faster than human-generated requests between January and May 2026. Simultaneously, corporate cost pressure is escalating; industry studies from McKinsey indicate that 93% of organizations are currently exceeding their allocated AI budgets. Fastly’s programmable edge infrastructure directly addresses these cost and security risks by applying real-time policy enforcement inline across public, private, and self-hosted model providers.
“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.”
Also Read: Abnormal AI Unveils AI Cloud Security Platform with OpenAI Daybreak Models
Three Core Critical Paths for Enterprise AI Governance
Fastly’s updated portfolio establishes centralized governance across the three primary operational vectors of modern AI systems: applications making model calls, users and external systems interacting with AI-powered applications, and autonomous software agents querying corporate APIs.
The newly released solutions include:
AI Runtime Control (Governing AI Access and Usage): Consolidates model calls through a single edge endpoint across public and self-hosted environments. Utilizing virtual keys to safeguard underlying provider credentials, the system offers real-time tracking of token spend, rate limiting, budget caps, and automatic failover routing to maintain model choice while capping runaway costs.
AI Firewall (Protecting AI Applications): Delivers inline threat protection at the edge to mitigate Large Language Model (LLM) vulnerabilities, including prompt injection attacks. By evaluating prompts directly within the request path before they reach downstream models, the firewall blocks malicious payloads deterministically without requiring GPU overhead.
API Security (Controlling Agent Interactions): Enforces strict API schemas and contract rules across agentic, AI-assisted, and conventional traffic. IT security teams can observe or block non-conforming requests on a service-by-service basis, preventing autonomous agents from initiating unsupported operations or transmitting malformed data.
Unifying Security and Edge Infrastructure
Built onto Fastly’s high-capacity global network which operates at 622 Tbps of capacity and processes over five trillion requests daily the new AI tools integrate seamlessly with the company’s existing security platform, including Next-Gen WAF, Bot Management, ContentGuard, DDoS Protection, and API Discovery. By embedding AI governance directly into the edge delivery path, Fastly enables enterprises to protect their digital assets, optimize model performance, and prevent unbudgeted expenditure from a single management console.