OpenAI's ChatGPT Work agent, powered by the newly approved GPT-5.6 model, targets enterprise productivity in a direct challenge to Microsoft Copilot and Google Gemini.
OpenAI's ChatGPT Work agent, powered by the newly approved GPT-5.6 model, targets enterprise productivity in a direct challenge to Microsoft Copilot and Google Gemini.

OpenAI's ChatGPT Work agent, powered by the newly approved GPT-5.6 model, targets enterprise productivity in a direct challenge to Microsoft Copilot and Google Gemini.
OpenAI's ChatGPT Work agent, capable of running complex business tasks for hours at a time, threatens to upend the enterprise AI productivity market dominated by Microsoft Copilot and Google Gemini. The tool, which launches Thursday alongside the GPT-5.6 model family, can create documents, spreadsheets, presentations and web applications autonomously, according to OpenAI.
"ChatGPT Work represents a meaningful step forward in what AI agents can accomplish independently over extended periods," a person familiar with the product strategy said. The agent is designed to handle multi-step workflows that previously required human supervision, the person added.
GPT-5.6 arrives in three tiers: Sol, the flagship model handling deep reasoning and agent orchestration; Terra, a mid-tier option balancing capability and cost; and Luna, the fastest and cheapest tier for routine tasks. OpenAI said Sol produces fewer factual errors than prior models and offers "a meaningful step up in cybersecurity capabilities," according to the company's deployment safety documentation. The release follows weeks of government review after the Trump administration last month pushed OpenAI to limit initial access to government-approved entities, delaying the full rollout.
The Commerce Department's Center for AI Standards and Innovation conducted additional testing before approving the broad release, with OpenAI sending technical experts to remain in Washington to address questions. The staggered rollout was not OpenAI's preferred approach, the company said at the time, noting that AI firms and the government are operating before more concrete standards for releasing frontier models — called for in President Trump's June 2 executive order — have been finalized. The restrictions echoed similar measures imposed on Anthropic, whose Mythos and Fable models were briefly banned from foreign access last month before the ban on Fable was lifted last week.
Who wins, who loses
The launch intensifies competition in the enterprise AI market, where Microsoft has embedded Copilot across its Office 365 suite serving more than 400 million paid seats, and Google has integrated Gemini into Workspace with over 3 billion users across its consumer and enterprise products. OpenAI's agent-based approach differs from both rivals by operating autonomously rather than as a side panel within existing productivity software, potentially appealing to companies seeking to automate entire workflows rather than assist individual tasks.
Microsoft shares fell 1.2% in pre-market trading Wednesday, while Alphabet shares slipped 0.4%, as investors weighed the competitive threat. OpenAI is not publicly traded, but the company's valuation has surged past $300 billion following its latest funding round, reflecting investor appetite for AI productivity tools.
Investment angle
For investors, the key question is whether ChatGPT Work can capture meaningful market share from entrenched competitors. Microsoft trades at 33x forward earnings, with Copilot revenue estimated at $12 billion annually by Morgan Stanley analysts. Google's Workspace contributes roughly $45 billion in annual revenue, though Gemini's direct contribution remains undisclosed. OpenAI's agent strategy could compress margins for both incumbents if it forces pricing competition, but the startup faces significant distribution challenges without an existing enterprise sales force or installed base.
This article is for informational purposes only and does not constitute investment advice.