How workers are unlocking new ways of working
- ID
- 25119
- Status
- summarized
- Published
- 16 Sep 2026, 5:00 PM
- Fetched
- 17 Sep 2026, 12:43 AM
- Provider
- OpenAI News
- Category
- ai-labs
- Original URL
- https://openai.com/index/unlocking-new-ways-of-working
- Source URL
- https://openai.com/news/rss.xml
Summary
- Score
- 5.5
- Created
- 17 Sep 2026, 12:44 AM
- Tags
- Audience
- ai-ml-learnersai-agent-userssaas-startup-founders
What happened
OpenAI analyzed 1.5M+ work-related ChatGPT messages (April–July 2026) and found workers repeatedly use AI for tasks outside their own occupation, turning one-off cross-occupation experiments into recurring workflow habits. Workers write shorter prompts for out-of-role tasks but are more likely to provide examples, background context, or ask AI to verify something—suggesting they use AI to 'borrow expertise' rather than learn a new field from scratch.
Why it matters
If you build AI tools or design team workflows, the pattern matters: users bring their own context (documents, examples) and want verification, not tutorials, when working outside their domain. Design for shorter prompts with rich attached context rather than long instructional prompts. For founders, this suggests AI adoption in orgs is less about training people in new skills and more about removing friction for cross-functional task execution.
Discussion angle
The 'borrowing expertise' framing vs 'learning a new field'—does this change how we design AI agent interfaces, onboarding, or team AI policies? Are we over-investing in teaching prompt engineering when users actually just want to attach a doc and get a domain-specific answer?