Managing humans is hard. Managing AI agents is… also hard. That’s why OpenAI is launching a new platform called OpenAI Frontier, which it says will help businesses “build, deploy, and manage” AI agents, even those not made by OpenAI itself.
OpenAI’s description of Frontier sounds something like HR for AI. “Frontier gives agents the same skills people need to succeed at work: shared context, onboarding, hands-on learning with feedback, and clear permissions and boundaries,” OpenAI wrote in a blog post. The similarity makes sense: OpenAI said the product was inspired “by looking at how enterprises already scale people.”
Frontier is available today, though only to an unspecified “limited set of customers, with broader availability coming over the next few months.” OpenAI said Intuit, State Farm, Thermo Fisher, and Uber are among the first companies to adopt OpenAI Frontier, with “dozens of existing customers” having piloted it as well. It’s not clear how much Frontier will cost, either. In a press briefing, chief revenue officer Denise Dresser declined to disclose pricing at this point in time.
Frontier is an “agent interface,” said Barret Zoph, OpenAI’s general manager for business-to-business, who recently returned to OpenAI after a stint at Thinking Machines Lab. Right now, many companies simply run AI agents on top of whatever they’re using, which often means fragmented tools, disconnected workflows, and siloed data. Frontier sits on top of that to create a “shared business context” for agents, connecting them with everything needed to work and communicate effectively. These connections mean deployed agents can operate across different environments, though OpenAI said Frontier also lets users set boundaries, making it “possible to use them confidently in sensitive and regulated environments.”
Frontier will also make it easy for human teams across organizations to “hire AI coworkers” for tasks like running code and data analysis. OpenAI said agents will also “build memories” and can be evaluated by human workers, which should make them more useful over time.
The end goal for OpenAI sounds suspiciously similar to Sauron’s motivation in The Lord of the Rings: one platform to rule them all. “By the end of the year, most digital work in leading enterprises will be directed by people and executed by fleets of agents,” said Fidji Simo, OpenAI’s CEO of Applications. “And what I dreamed of was having one platform to create and manage all of our agents.”
Interestingly, this means “a recognition that we’re not going to build everything ourselves,” said Simo. Frontier will use open standards and can be populated with agents built by OpenAI, the enterprise customer, or another AI company.
Frontier comes as AI companies strive to prove AI tools are genuinely useful for their customers, and work to create revenue streams that justify the enormous amount of money being pumped into the sector. Agents, tools that can act independently, are a core focal point for this, and Frontier can be seen as a direct response to Microsoft’s Agent 365 agent manager. Anthropic is strong competition too, after its Claude Cowork and Claude Code took the AI industry by storm.
