Oracle's recruiters spent up to 4 days on research. ChatGPT made it 20 minutes.
The figures come from a customer story that OpenAI published on 8 October. Oracle's talent acquisition team built an internal tool on ChatGPT Work. A recruiter gives it a job description, and it looks up comparable roles, pay benchmarks and the talent pool in each location.
Compiling that used to take two to four days before each search, the story says. "Now we're able to sit down and prep for about 15 to 20 minutes," said Jan Ackerman, Oracle's global head of talent acquisition. According to the story, the tool also standardises the research recruiters bring to hiring managers.
OpenAI says more than 100,000 Oracle employees use ChatGPT Work and Codex. The story also reports that Codex cut resolution time for simple IT incidents from about an hour to minutes.
The time saved is on research before a search, not on the whole hiring process. The story does not say which sources feed the pay benchmarks, or how recruiters check them before advising a hiring manager. It gives no count of recruiters using the tool. These are the two companies' own numbers.
🎥 Video: OpenAI, official YouTube channel · narration: AI voice
What the captions say
- Oracle built a recruiting tool on ChatGPT.
- In: a job description. Out: roles, pay, talent.
- That research used to take 2–4 days.
- Now recruiters prep in 15–20 minutes.
- Every hiring manager gets the same data.
- 100,000+ Oracle staff use ChatGPT or Codex.
- Figures are OpenAI's and Oracle's own.