OpenAI has acquired cloud-based developer Ona to power Codex with secure, customer-controlled cloud infrastructure for long-running AI agents. The ChatGPT parent company said the integration will allow Codex to complete tasks that could take hours or days in minutes, while also making it easier for organizations to deploy agents in production with stronger visibility and oversight. The transaction remains subject to customary closing conditions and regulatory approvals, and the companies will continue operating independently until the deal is completed.

Ona, which has helped shift software development workflows into cloud-based environments and supports more than 2 million developers, already shares customers with OpenAI. The acquisition comes as Codex expands beyond its roots as a coding assistant into a broader platform for research, analysis, automation and software development. OpenAI said Codex now has more than 5 million weekly users, a 400% increase from earlier this year.

Strategic Integration

OpenAI, which is backed by Microsoft Corp (NASDAQ: MSFT), said organizations should be able to use persistent, agent-based systems with confidence, knowing those agents run within environments that satisfy their security, governance, and operational standards. The company added that this requires clear control over deployment environments, system access, credential boundaries, activity logging, and the review process for how work is executed and approved.

Leadership Perspective

"Agents need more than intelligence; they need a trusted workspace," Ona co-founder and Chief Executive Johannes Landgraf said in a statement. "We built Ona to give agents cloud environments with the context, control and collaboration enterprises require. Joining OpenAI lets us bring that foundation into Codex, helping organizations deploy agents with confidence and giving humans more agency over their work."

Future Operations

After the acquisition closes, Ona’s team will join OpenAI and work alongside the Codex group to build secure, persistent execution capabilities for enterprise customers. Together, the two companies aim to help engineering teams take on long-running software tasks more safely and effectively across the full lifecycle—from running tests and fixing issues to modernizing applications, patching vulnerabilities, and managing complex workflows over time.

Former hedge fund manager Martin Shkreli predicted Thursday that Meta Platforms Inc. will absorb Anthropic in a "down exit." The forecast arrives as intense public backlash, regulatory hurdles, and a looming IPO test the artificial intelligence startup's trillion-dollar ambitions. Sharing on X, Shkreli floated that Anthropic's current trajectory could force an unfavorable Meta acquisition, though he caveated the statement as a "low conviction" prediction. The commentary underscores mounting skepticism regarding Anthropic's ability to sustain its independence in a cutthroat AI arms race.

Mounting Backlash and Operational Pressures

Shkreli's prediction amplified a warning from a tech commentator on X, who told Anthropic employees that leadership's recent policy decisions are "titanic f***ups that pose serious risks to both your technical pole position and your bags." Users are increasingly frustrated over suspected "stealth nerfing" of Claude's capabilities and a controversial update extending data retention to five years for model training. The operational strain has grown complex. The Pentagon recently slapped Anthropic with a supply chain risk designation, temporarily blocking new Defense Department contracts. Additionally, leadership recently called for an industry-wide development pause, sparking debate over whether their strict safety boundaries are actively dampening their competitive edge.

The Trillion-Dollar IPO Gamble

Paradoxically, this turbulence collides with Anthropic's monumental push to go public. On June 1, the company filed a confidential draft S-1 with the SEC, fresh off a $65 billion Series H round that valued the startup at a staggering $965 billion. While Anthropic boasts an explosive $47 billion annualized revenue run rate and massive infrastructure deals with Amazon.com Inc. and Alphabet Inc.'s Google, its impending market debut faces heavy headwinds. The company will have to publicly justify its previously shielded margins and cost structures. Furthermore, it is stepping into an overwhelmingly crowded 2026 IPO window, forced to compete directly for institutional capital against massive upcoming listings from OpenAI and SpaceX. If public investors balk at the nearly trillion-dollar price tag, Shkreli's discounted acquisition scenario could quickly become a reality.