Kuken Industries uses OpenAI API to search decades of drawings
Kuken Industries has applied the OpenAI API to its archive of cooling tower drawings, letting engineers locate the right document in minutes instead of close to an hour.

Key takeaways
- Kuken Industries uses the OpenAI API to search 70 years of cooling tower drawings.
- Engineers can locate a repair drawing in 4.5 minutes, according to OpenAI.
- The same search previously took close to an hour.
- The results come from OpenAI's own announcement and have not been independently verified.
In this article
What the company built
Kuken Industries has connected the OpenAI API to its archive of cooling tower drawings, according to OpenAI News. The archive spans 70 years of engineering documents.
The system lets engineers search that collection by asking for what they need, rather than browsing manually.
Time saved on repairs

OpenAI reports that an engineer can now find the drawing needed for a repair in 4.5 minutes. Before the system was introduced, the same task took close to an hour.
That reduction matters because crews often need the correct drawing before repair work can begin.
Why it matters
Long-standing industrial firms often hold decades of paper or scanned records that are hard to search. This case shows how large language models can make such archives usable in day-to-day work.
The figures come from OpenAI's own announcement, so they reflect the company's description of the deployment rather than independent testing.
Sources
Spotted a mistake? See our corrections policy. Read how News by OnDevice works in our editorial policy.



