NTT DATA Group cuts incident analysis to 30 minutes with Codex
OpenAI
NTT DATA rolled out ChatGPT Enterprise and Codex to 9,000 staff, and incident analysis that used to eat hours now takes 30 minutes. That's a huge win for a company whose whole business is enterprise IT reliability.
Based on reporting by OpenAI — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
NTT DATA Group isn't a scrappy startup experimenting with AI on the side. It's one of the largest IT services firms on the planet, the kind of company that runs infrastructure and support desks for other massive corporations. So when it says a core operational task now takes 30 minutes instead of whatever it used to take, that's not a marketing flourish, that's a real shift in how a huge workforce spends its day.
The tool doing the heavy lifting here is Codex, OpenAI's coding agent, paired with a company-wide rollout of ChatGPT Enterprise across roughly 9,000 employees. Incident analysis, the unglamorous but critical work of figuring out why something broke and how to fix it, is exactly the kind of task that eats analyst hours and delays resolution. Getting that down to half an hour changes the math on how many incidents a team can actually work through in a day.
What's notable is the scale of adoption rather than a flashy pilot program. Nine thousand employees is not a lab experiment, it's a genuine attempt to bake AI into daily workflows across a company whose entire reputation rests on operational trust. NTT DATA has spent decades selling reliability and security to enterprise clients, so handing that many staff generative AI tools required them to also be confident about governance and data handling, not just productivity gains.
The broader pattern here is one we're seeing repeat across large Japanese and global systems integrators: less excitement about chatbots writing marketing copy, more focus on AI quietly compressing the grunt work inside IT operations. Incident response, ticket triage, code review, these are the places where minutes saved multiply across thousands of employees and translate into real cost and speed advantages. NTT DATA's numbers are a data point, but they're also a signal of where enterprise AI spending is actually going.
My take — AI-written commentary, not fact-checked reporting
This is the boring-but-important side of AI that doesn't trend on social media, and that's exactly why it matters. Nobody's tweeting about incident analysis times, but shaving hours down to 30 minutes across 9,000 employees is the kind of unglamorous productivity gain that actually shows up in a P&L, unlike a lot of the flashy demo-ware we cover here.
Read more about this at: OpenAI
