My IBM Log in

IBM API Connect expands its AI Gateway feature to additional models and deployments

13 November 2024

 

Author

Aaron Lieber

Program Director, Product Management - API Connect

IBM

Generative AI (GenAI) is revolutionizing how organizations unlock rapid innovation and drive massive productivity gains. However, the swift pace of adoption often leaves business leaders grappling with a lack of visibility and control over their enterprise's use of GenAI. According to Gartner, over 80% of organizations will have deployed GenAI applications or used GenAI APIs by 2026, yet the IBM Institute for Business Value reports that 60% of organizations have not yet developed a consistent, enterprise-wide approach to GenAI.

Business leaders often find themselves asking:

  • Which AI services are being used across my organization?
  • How are they being used?
  • How much are they cumulatively costing us?
  • How can I ensure they’re used securely?

If you struggle to answer these questions, an AI gateway can help.

What is an AI gateway?

An AI gateway provides a single point of control for organizations to access AI services via APIs in the public domain and brokers secured connectivity between your different applications and third-party AI APIs both within and outside an organization’s infrastructure. It acts as the gatekeeper for data and instructions that flow between those components. An AI gateway provides policies to centrally manage and control the use of AI APIs with your applications, as well as key analytics and insights to help you make top-level decisions faster on which AI services to prioritize. Implementing an AI gateway can provide the centralized visibility needed to harness GenAI securely and effectively.

Announcing AI Gateway for IBM API Connect: Now available on-prem

In June, we introduced the AI gateway feature in IBM API Connect on AWS, our SaaS offering hosted on Amazon Web Services. Today, we are excited to announce the availability of the on-premises version of the AI Gateway, which is now available to all API Connect customers with the installation of version 10.0.8.1. With this latest release, you can manage and secure your AI services APIs across your cloud and on-prem applications.

The initial launch of the AI Gateway in June provided support for watsonx.ai APIs, and we promised additional AI services would be added later in the year. This latest version of the AI Gateway adds out-of-the-box support for OpenAI’s ChatGPT, further empowering customers to accelerate their AI journey by centrally managing both watsonx.ai and ChatGPT APIs. Stay tuned in the coming months for even more pre-built policies supporting additional AI services.

Key benefits of AI Gateway for IBM API Connect

Faster and more responsible adoption of GenAI

A guided wizard walks the user through configuration parameters, making it easy for developers to get self-service access to enterprise AI APIs, accelerating secure and responsible adoption of GenAI enterprise-wide.

Improved insights and cost management

Use built-in analytics and dashboards to get visibility into organization-wide use of AI APIs. With these analytics, you’ll be able to answer the questions:

  • Which AI services are being used across my organization?
  • How are they being used?
  • How much are they cumulatively costing us?
  • How can I ensure they’re used securely?

In addition, with the AI gateway, you can prevent unexpected or excessive costs for AI services by limiting the request rate within a specified duration and by caching AI responses.

Enhanced governance and compliance

By funneling all AI services traffic through the AI Gateway, you can centrally manage the use and compliance of AI services through policy enforcement, data encryption, masking of sensitive data, access control, audit trails and more.

Take the next step

Get more transparency, cost savings, and control of your AI services with API Connect’s AI Gateway.

Visit API Connect's AI Gateway website.

Sign up today for a demo of API Connect.