Secure, trustworthy, enterprise-ready AI

Turnkey AI for the enterprise

AI platforms are bringing business outcomes and ROI to the enterprise faster, often with less cost and complexity. Nutanix Enterprise AI is a turnkey, full-stack, software-defined AI solution that accelerates the deployment of AI models and inferencing quickly and at scale—on premises, in the cloud, or at the edge. It includes a pre-validated, end-to-end platform that is fully integrated, fast to deploy, and simple to manage.

At its core, Nutanix Enterprise AI includes two key components:

1. The AI Agent Gateway, which provides centralized cost governance, controlled access to models and tools, and unified observability over token usage
2. The Private Inference offering, featuring a curated model catalog, optimized LLM inference, and enterprise-grade auditability

Even turnkey AI platforms require enterprise-grade security and governance. As businesses become more dependent on AI workflows, protecting applications, data, and models—and demonstrating compliance with evolving AI regulations—becomes a business imperative.

F5 and Nutanix for AI application delivery and security

Addressing AI risk requires continuous command and connected defenses with the capability to define, observe, evaluate, and enforce controls tailored to unique risk-definitions and business needs.

The F5® Application Delivery and Security Platform (ADSP) provides the critical services needed to ensure AI applications, agents, and models stay secure and compliant. By optimizing AI traffic flows across distributed environments, the F5 ADSP improves performance predictability, increases GPU usage, simplifies operations, and provides adaptive runtime AI security.

Built-in resilience for AI inference

To ensure better availability and control, Nutanix Enterprise AI provides a native, centralized traffic controller for LLM inference. Organizations get consistent authentication, rate limiting, and usage reporting in one centralized platform for better cost control and governance.

The Nutanix AI Agent Gateway handles all model load balancing, token-based rate limiting, and model fallback while delivering unified observability, so organizations can see who is using AI models and how tokens are being consumed. The gateway ensures that apps, models, and agents remain available and usage costs are controlled.

However, AI applications may still be vulnerable to AI-specific attacks like prompt injections or jailbreaks. That is where F5 AI Guardrails comes in.

Guardrails for Nutanix Enterprise AI applications

Ensuring responsible AI usage requires granular controls that can be applied efficiently to user-initiated AI prompts and AI outputs. From protecting applications, models, and data from increasingly sophisticated AI threats to keeping AI usage aligned with security and compliance policies, an additional layer of protection is required.

F5 AI Guardrails sits between end-user-facing AI applications and backend inference systems, intercepting, monitoring, and blocking attacks. This runtime protection reduces AI risk, minimizes your attack surface, and ensures AI governance policies are continually and consistently enforced.

Figure 1: AI Guardrails provides real-time protection for Nutanix Enterprise AI by securing LLM traffic against AI-specific threats.

AI Guardrails applies built-in and customizable scanners to inspect and filter LLM inputs and outputs, helping prevent harmful, non-compliant, or inappropriate responses that could create liability for the business or potential harm to end users. Out-of-the-box scanners cover commonly restricted topics—including inquiries for medical, financial, or legal advice. These scanners can also be customized using natural language to address your unique use cases.

Figure 2: AI Guardrails contains out-of-the-box guardrails around commonly restricted topics.

Scale enterprise AI with confidence

Together, F5 and Nutanix speed the adoption of enterprise AI by delivering an easy-to-deploy, end-to-end solution that increases both the resiliency and security of AI use, while ensuring token consumption, inputs, and outputs remain compliant with an organization's policies.

Learn more about secure AI solutions from F5 and Nutanix at f5.com/nutanix.

KEY BENEFITS

Protect AI apps, data, and models

Establish adaptive security that blocks evolving AI attacks including jailbreaks, prompt injection, and data leakage

Centralized observability and control

Gain a granular, centralized view of token usage, control consumption with rate limiting, and optimize with local model routing

Accelerate secure AI deployment

Stand-up turnkey enterprise AI solutions and protections on any CNCF Kubernetes distribution, including F5 AI Guardrails and the Nutanix Kubernetes Platform

Scale AI confidently

Together, F5 and Nutanix provide cost-controlled, compliant, and secure AI operations