Uber brings ChatGPT Work to all 30,000 employees as its first company-wide AI agent platform

Uber is rolling out ChatGPT Work to its entire global workforce of more than 30,000 employees, OpenAI announced on October 22, 2026. The expansion scales a pilot that started with 1,000 employees earlier this year and makes ChatGPT Work the company's first company-wide deployment of an agentic platform for knowledge work. Uber's CTO Praveen Neppalli Naga framed it as giving everyone tools to build, while OpenAI president Greg Brockman called it a step toward broader access to frontier models and agents.

ChatGPT Work at Uber: from 1,000-person pilot to all 30,000 employees

ChatGPT Work, introduced in July 2026, is the agent inside ChatGPT Enterprise that takes on multi-step tasks and returns finished work rather than chat replies. At Uber it connects to Google Drive, Gmail, Google Calendar and Slack, so employees can research, prepare meeting briefs, build presentations, analyze data and run cross-app workflows from one place. TechCrunch's report adds detail from Uber's side: roughly 80% of the original pilot users were non-engineers, and they used the agent to analyze operations data, localize training materials across markets and languages, and build personal interactive dashboards.

GPT-6.1 Sol becomes the default inside ChatGPT Work

The announcement puts OpenAI's newest models directly behind the enterprise agent. GPT-6.1 Sol, introduced on September 29, 2026, is deployed as the default capability in ChatGPT Work, which OpenAI describes as balancing frontier performance with cost efficiency. Uber teams also get multimodal voice and image generation, and their engineering organizations keep using Codex for software work. For readers already on ChatGPT Enterprise, the Uber deployment signals where OpenAI is steering the platform: an agent layer with model defaults and app connectors, not just a chat box.

Uber's rollout plan: skills library, AI pods and outcome tracking

The rollout comes with Uber's internal AI skills library, approved integrations and structured employee training across operations, finance, marketing and product. Uber's VP of engineering Sumanth Sukumar told TechCrunch that teams earlier paired 30 AI engineers with business units for two-week "AI pods" to automate key parts of jobs. Uber declined to share pricing or licensing terms, but says it will track how employees use the tool and which workflows it improves. The two companies plan to publish adoption insights and measurable outcomes as the rollout progresses.

Why the enterprise agent race is the real story here

TechCrunch positions the move as part of OpenAI's rapid enterprise push against Anthropic, a market that other AI companies have struggled to serve at this scale. Watch a few concrete signals: whether OpenAI names ChatGPT Work in its release notes and plan pages, whether other large employers announce comparable company-wide agent deployments, and whether the promised Uber adoption data shows measurable workflow gains. The pattern behind this deal is larger than one ride-share company: agentic platforms are becoming the default interface for office work, and they are now sold to entire companies rather than to individual teams.

Practical takeaway for teams evaluating this shift: map the connector surface (Drive, Gmail, Calendar, Slack) against your own stack, pilot with non-engineer functions first, and set up usage and outcome tracking before scaling seats. If you are building agent workflows yourself, our guide to AI agent tools and AI workflow automation covers the current landscape.

Sources and caveats

Primary evidence is OpenAI's official announcement; independent reporting and additional detail (80% non-engineer pilot share, AI pods, undisclosed pricing) come from TechCrunch. Uber's rollout figures and model defaults are the companies' own statements and have not been independently measured. Adoption outcomes are promised but not yet published, so any efficiency claims should stay unverified until data lands.