Compass Investments

OpenAI has unveiled a fresh enterprise platform called Frontier, designed to help organizations integrate AI agents into their systems.

Frontier connects back-end programs, data and utilities to enable agents to perform real-world workloads within companies.

The platform supports the use of agents from OpenAI, developed in-house by teams, as well as third-party vendors such as Google and Microsoft.

OpenAI has released a new platform called Frontier, aimed at companies that want to leverage artificial intelligence without disrupting established infrastructure. The platform functions by aggregating heterogeneous tools, applications and company information into a single space. This provides AI agents with the information they need to complete tasks within the organization.

These agents are able to act autonomously without human intervention, activating utilities, generating code, managing files and more. OpenAI says the goal is to provide businesses with a unified and easy-to-understand method of creating, managing and running agents that actually benefit their operations.

It’s basically a recognition that we don’t intend to build everything ourselves, shares Fiji Simo, head of applications at OpenAI. We’re going to partner with the ecosystem to build together, and we accept the fact that enterprises are going to need a lot of different partners.

OpenAI has made it clear that this is not a one-size-fits-all solution. Frontier can work with agents built by OpenAI, but it is also compatible with custom agents developed by user companies. It even works with third-party agents from Google, Microsoft, and Anthropic.

According to Simo, OpenAI is simply not physically capable of creating all the AI agents that companies need.

This platform does not completely replace tools that are already available. It integrates with those that organizations already use, including ChatGPT Enterprise.

What makes it different is that it helps agents function like real team members within the organization. They gain access to the overall business context of the company by connecting to resources such as application systems, internal applications, and databases.

What most firms are really missing is a simple way to unlock the potential of agents as colleagues who can work within the business without having to be completely rebuilt for them, said Denise Dresser, chief revenue officer at OpenAI. “That’s why we developed Frontier.

It also includes mechanisms for monitoring agent performance and improving it later. These integrated utilities are designed to test, evaluate, and improve performance on a variety of tasks so agents don’t sit idle.

Barrett Zoph, OpenAI’s general manager for enterprise, said the company is committed to “turning agents into full-fledged AI employees.

He joined OpenAI in January after leaving Thinking Machines Lab, a startup he co-founded with former OpenAI CTO Mira Murati.

Frontier is currently available to a limited number of early adopters. These include Uber, State Farm, Intuit, and Thermo Fisher. Wider access is expected over the next few months, according to OpenAI. Pricing information was not disclosed.

This new product launch follows an obvious trend. In November, OpenAI reported that its technology is used by more than a million businesses. According to CFO Sarah Fryar, enterprise customers already account for about 40% of the company’s turnover, and she predicts that share will reach 50% by the end of 2026.