OpenAI acquires Ona to make Codex enterprise-ready with secure cloud environments

Reviewed byNidhi Govil

9 Sources

Share

OpenAI announced plans to acquire Ona, formerly known as Gitpod, to strengthen Codex cloud capabilities and address enterprise security concerns around AI agents. The 79-person company provides secure cloud environments where AI agents can run long-running tasks that span hours or days, even when developers are offline. With Codex usage surging 400% and now serving over 5 million weekly users, the acquisition positions OpenAI to compete directly with Anthropic's Claude Code in the race for enterprise AI adoption.

OpenAI Acquires Ona to Address Enterprise AI Agent Concerns

OpenAI announced on Thursday that it has agreed to acquire Ona, a 79-person cloud development environment provider formerly known as Gitpod, marking a strategic move to make AI agents more palatable to enterprise customers

1

. The acquisition directly addresses the fears that CIOs and CISOs harbor about deploying fully autonomous AI agents in production environments—concerns ranging from agents deleting critical files to generating massive token bills or being exploited by malicious actors

1

. OpenAI did not disclose the financial terms of the deal, which remains subject to customary regulatory approvals

2

.

Source: ET

Source: ET

The acquisition will integrate Ona's technology into Codex, OpenAI's AI coding assistant that has experienced explosive growth. More than 5 million people now use Codex each week, representing a 400% increase since early this year

3

5

. This surge reflects how AI coding assistants have become mainstream tools for software developers, with some competitors in the space reaching $500 million in revenue

3

.

Source: InfoWorld

Source: InfoWorld

Secure Cloud Environments Enable Long-Running Tasks

Ona's platform provides the infrastructure that Codex needs to handle increasingly complex workflows. The company offers secure, persistent cloud environments—essentially sandboxes—where AI agents can continue working even after developers close their laptops

3

4

. This capability is critical as the work that Codex performs has evolved from tasks taking minutes to long-running tasks that stretch across hours or even days

3

5

.

When developers run AI agents on local machines, those agents stop working when the workstation powers down, fundamentally limiting their utility for extended projects like modernizing an entire codebase or patching vulnerability classes across multiple systems

4

. Ona's sandboxes remain online continuously, ensuring uninterrupted operation

4

. Additionally, the cloud infrastructure offers access to greater computing resources than typical developer workstations, enabling agents to complete tasks more quickly

4

.

Customer-Controlled Execution Builds Enterprise Trust

The real value proposition centers on control rather than raw performance. Ona's approach to customer-controlled execution allows AI agents to run inside a company's own cloud infrastructure

3

5

. OpenAI provides the intelligence, but customers retain ownership of their data, credentials, and audit trails—a trust pitch aimed squarely at risk-averse IT departments

3

. "Agents need more than intelligence; they need a trusted workspace," said Ona co-founder and CEO Johannes Landgraf

3

.

Ona's platform includes multiple cybersecurity guardrails that make enterprise-friendly AI agents feasible. Administrators can configure sandboxes to prevent agents from accessing risky programs, with the system capable of detecting blocked applications even when hackers rename them, move them to different file paths, or hide them in scripts

4

. This detection relies on hashing, which generates unique cryptographic signatures for programs that administrators wish to block

4

. The platform also prevents AI agents from accessing file system areas containing sensitive credentials like encryption keys and blocks outbound connections to potentially malicious servers

4

. Crucially, Ona deletes sandboxes when they're no longer needed, reducing data theft risks compared to local AI agent environments that often persist longer than necessary

4

.

Racing Anthropic for Enterprise Market Share

This acquisition represents a direct competitive response to Anthropic, whose Claude Code has helped drive a year of explosive growth for the AI startup

3

. Both companies are racing to become the AI agent orchestration platform that large enterprises trust with production systems

3

. The timing carries additional weight: OpenAI filed confidentially for an IPO on Monday, just days after Anthropic did the same

3

. For both companies, Codex revenue and enterprise credibility feed an investor narrative as they prepare for public markets

3

.

Ona itself is not a newcomer to the developer tools space. The company began as Gitpod, a German startup that pioneered moving coding work from local machines into the cloud, serving approximately 2 million developers over its history

3

5

. It rebranded to Ona in late 2025 and rebuilt its platform specifically around AI agents before being absorbed by the US giant

3

. "I always thought selling the company would feel like an ending," Landgraf wrote on LinkedIn. "Instead, it feels like our life's work just got bigger and more important"

2

.

Part of Broader Acquisition Strategy to Strengthen Codex Cloud Capabilities

The Ona deal fits into a larger acquisition pattern at OpenAI. The company acquired Promptfoo Inc. in March 2026 to enhance cybersecurity features of its OpenAI Frontier platform, which enables companies to build custom AI agents

4

. Promptfoo developed tools that detect and fix vulnerabilities in AI applications

4

. Last year, OpenAI also picked up Jony Ive's $6 billion hardware venture, io

3

. Each acquisition plugs a specific gap, but the Ona deal addresses what may matter most for selling agents to enterprises: running them where customers feel safe

3

.

Source: PYMNTS

Source: PYMNTS

Once the deal closes, Ona's team will join OpenAI and work directly on the Codex team to implement these upgrades

2

4

. The acquisition will provide Codex users with self-hosted sandboxes, moving tool execution into infrastructure that users control

1

. This shift enables users to review the work of agents and provide input when necessary during extended operations

4

. Until regulatory approval is secured and the deal officially closes, the two companies will continue operating independently

3

5

. The direction, however, is clear: OpenAI wants Codex deployed everywhere serious enterprise-ready work happens, and it just acquired the plumbing to make that vision achievable [3](https://thenextweb.com/news/openai-acquires-ona-codex].

Today's Top Stories

© 2026 TheOutpost.AI All rights reserved