OpenAI has agreed to acquire Ona, a cloud infrastructure startup focused on providing secure environments for AI agents, a move aimed at expanding the capabilities of its Codex coding platform for enterprise users.
OpenAI said more than 5 million people now use Codex weekly. The company also reported that usage has increased roughly 400% since the beginning of the year, highlighting growing demand for AI systems capable of handling software development and other complex workflows.
A major challenge has emerged as AI agents take on larger assignments. Rather than completing tasks in minutes, newer systems are expected to work for hours or even days. That requires persistent execution environments that remain active after a user ends a session.
Ona specializes in that capability. Its platform provides cloud-based environments where AI agents can access tools and contextual information over extended periods while maintaining security controls. The technology allows work to continue without depending on a developer to keep an active connection.
The company is also known for enabling what it describes as customer-controlled execution. Under that model, agents can operate inside a customer’s own cloud infrastructure rather than within a third-party environment. Customers retain control over data, credentials, access policies, audit logs, and governance requirements while still benefiting from AI-driven automation.
That feature is particularly significant for large enterprises, where security, compliance, and operational oversight often determine whether AI projects move from pilot programs into production.
Ona traces its roots to Gitpod, a German developer tools company that helped software teams move development workflows into cloud environments. The company says its technology has been used by approximately 2 million developers. It later repositioned itself around infrastructure designed specifically for AI agents and rebranded as Ona.
Battle with Anthropic
OpenAI is engaged in a high-stakes battle with Anthropic, and both AI leaders have confidentially submitted IPO-related filings, starting a new phase in the race to capture enterprise AI spending.
As part of that competition, both companies are investing heavily in AI coding assistants and enterprise-focused agent platforms. Coding tools have become one of the fastest-growing categories within generative AI, with businesses investing in systems that can write code, test applications, resolve software issues, and automate development tasks.
The Ona acquisition continues an active acquisition strategy at OpenAI. In recent months, the company has purchased cybersecurity startup Promptfoo and previously acquired several companies focused on AI interfaces, healthcare tech, and hardware development.
Unlike some earlier acquisitions that targeted user experiences or specialized applications, Ona addresses a foundational infrastructure requirement. As AI agents become more autonomous and are entrusted with ever more important business tasks, companies need greater control over where those agents operate and how they access sensitive systems. Governance has moved to the forefront.
Financial terms were not disclosed, and the transaction remains subject to customary closing conditions and regulatory approvals. Once completed, Ona’s team will join OpenAI and work within the Codex organization.
