OpenAI launched a powerful new enterprise platform, called Frontier, in a move designed to integrate artificial intelligence (AI) agents directly into the existing software ecosystems of global corporations.
Thursday’s announcement, which positions AI agents as autonomous digital teammates rather than just chatbots, sent ripples through the tech sector, causing a notable dip in the stocks of traditional software giants.
Frontier is designed to serve as an intelligence layer that bridges the gap between siloed data sources. Rather than requiring companies to abandon their current software, Frontier connects disparate systems — such as CRM platforms, data warehouses, and internal ticketing tools — to provide AI agents with what it calls “shared business context.”
According to OpenAI executives, the goal is to move beyond isolated use cases toward true AI co-workers. These agents can reason over complex data, run code, and manage files autonomously within a secure execution environment.
“Frontier is really a recognition that we’re not going to build everything ourselves,” said Fidji Simo, OpenAI’s CEO of Applications. “We embrace the fact that enterprises are going to need a lot of different partners.”
The platform arrives as OpenAI aggressively scales its business-to-business (B2B) operations. OpenAI Chief Financial Officer Sarah Friar recently noted that enterprise customers now account for approximately 40% of the company’s revenue, a figure expected to hit 50% by year-end.
“The AI experimentation phase is now fully in our rearview mirror. OpenAI Frontier signals that AI development is entering a phase where frontier capability is managed as governed infrastructure,” said Mitch Ashley, vice president and practice lead, Software Lifecycle Engineering, at The Futurum Group. “The emphasis is on controlled access, evaluation rigor, and shared responsibility for agent-based systems. That framing reflects a maturing view of AI as something that must be operationalized with intent, constraints, and accountability from the outset.”
“For software lifecycle engineering, this reinforces a broader pattern already playing out across agent-driven development,” Ashley added. “As AI systems move closer to execution authority, platforms must provide clear boundaries around who can do what, when, and under what conditions. Frontier is another marker that the industry is shifting toward AI systems designed to be deployed, observed, and governed in production environments.”
OpenAI’s agentic AI bid comes amid a hiring binge of hundreds of AI consultants to boost enterprise sales, according to a report in The Information.
A heavy-hitting roster of early adopters has already signed on with OpenAI, including Uber Technologies Inc., Oracle Corp., HP Inc., State Farm Insurance, Intuit Inc., and Thermo Fisher Scientific. Meanwhile, Cisco Systems Inc. and T-Mobile U.S. Inc. have successfully piloted the technology.
OpenAI claims that more than 1 million businesses are now using its broader software suite, reporting dramatic efficiency gains, including a manufacturer that reportedly reduced production optimization timelines from six weeks to a single day.
The news hit the software market hard, cutting short a brief recovery for several industry incumbents. Salesforce Inc. shares dropped more than 3%, while Thomson Reuters fell 4.6%.
Investor anxiety stems from the rapid evolution of agentic AI. The fear is that if companies can build bespoke AI agents to handle marketing, legal, and sales workflows via Frontier, the need for traditional, high-cost SaaS (Software-as-a-Service) subscriptions could evaporate.
However, some industry leaders remain skeptical of total displacement. NVIDIA Corp. CEO Jensen Huang recently dismissed the idea that traditional software is doomed, arguing that most businesses prefer to enhance their existing tools rather than “completely reinvent” their digital infrastructure.
Frontier is initially launching to a select group of enterprise clients, with broader availability expected in the coming months. While OpenAI declined to disclose specific pricing, the company emphasized that the platform is compatible with third-party agents from rivals like Google, Microsoft Corp., and Anthropic, signaling a push for Frontier to become the central operating system for the modern AI-driven enterprise.
