NTT DATA Group cut a complex incident analysis that used to take three days with five engineers down to just 30 minutes using Codex, the advanced automation capability tied to ChatGPT Enterprise. The company already has around 9,000 employees using Codex and is transforming how work gets done across both technical and non-technical teams.
What NTT DATA did and why it matters
The Japanese firm rolled out ChatGPT Enterprise across the organization and created an OpenAI Center of Excellence (CoE) to ease adoption, governance, and best practices. The result: everyday AI use, more confidence to delegate tasks, and Codex adoption that already reaches thousands of employees.
Why is this relevant to you? Because it’s not just about speeding up code: it’s about delegating well-defined tasks to an assistant that investigates, executes, tests, and reviews on its own within safe boundaries.
In an internal survey more than 96% said they were satisfied with ChatGPT Enterprise and over 95% reported productivity gains.
The case that caught attention: 3 days to 30 minutes
Before: five engineers spent three days analyzing a critical incident. Now: Codex completed the process in 30 minutes. How can that happen? Codex doesn't just suggest lines of code; it can orchestrate an investigation: gather data, run automated tests, analyze results, and adjust steps.
This success was the first proof point that convinced leadership and helped drive adoption across the organization.
How non-technical employees use Codex
Codex didn’t stay only with the developer team. Under the 'Client Zero' philosophy (test internally before taking it to clients), finance, operations, and service teams are building practical flows:
- Extract travel expenses from bank statements and populate expense forms.
- Organize and summarize large volumes of documents for audits or reviews.
- Analyze data in Excel and generate reports without building a full BI solution.
- Automate repetitive tasks with Playwright and package them as 'Skills' for company-wide use.
Result: tasks that used to need technical support or specialized tools can now be done by employees with functional knowledge.
Security and governance: the foundation for using Codex in the enterprise
NTT DATA understood that to scale they needed clear rules. The CoE defined guidelines that state:
- what data can be used;
- which systems Codex can connect to;
- how to manage network traffic and what sandboxing mode to apply;
- what level of automation is appropriate and where human review must always occur.
That security layer let more people explore without fear of exposing sensitive information and fostered internal trust.
Measurable results
- Complex incident analysis completed in 30 minutes (before: 5 engineers, 3 days).
- Approximately 9,000 employees with access to and use of Codex.
- 1.4× increase in weekly active users after publishing a usage guide and running hands-on training.
- Internal automations with Playwright deployed as Skills for organizational use.
Five practical lessons for other companies
- Make ChatGPT Enterprise a day-to-day tool to build habits of collaborating with AI.
- Deploy widely to leverage network effects (peer learning and word of mouth).
- Implement clear security and privacy before scaling usage.
- Treat the rollout as the beginning: continuously improve with usage data, surveys, and interviews.
- The CoE should identify high-impact cases and turn them into reusable processes with safeguards.
Impact and vision: AI as a workmate
NTT DATA isn’t trying to replace people but to amplify their capabilities. The idea is simple: people set direction and evaluate outcomes; AI interprets information, executes defined steps, and speeds up value delivery. When this reaches both technical and non-technical employees, the organization gains speed and reduces dependence on specialists for routine tasks.
Can you imagine your team freeing up hours every week and focusing on what truly adds value? That’s the leap NTT DATA is already experiencing.