How workers are unlocking new ways of working

New OpenAI Economic Research shows how workers use AI to do tasks outside their typical occupation and which new tasks they build into their workflow.

Our latest research suggests the importance of how tasks are divided and carried out alongside access to AI tools. These new tasks outside their typical job descriptions become a regular part of how workers use AI. If those activities become regular responsibilities, jobs could broaden even while their titles stay the same.

Our first Work at the Frontier report⁠ documented ‘task crossover’: workers using AI for activities historically associated with another occupation. Our latest research asks what happens next.

Analyzing more than 1.5 million work-related ChatGPT messages from April through July 2026, we find evidence that some cross-occupation AI use is recurring. Workers return to tasks outside their occupational boundaries, and these activities become a larger part of their observed AI use over time.

The findings offer an early look at how access to AI may translate from trying new tasks to incorporating those tasks into workers’ regular routines. For companies that are endeavoring to determine how best to adopt and make use of AI in their organizations, the trends identified in our four-month study of professional worker AI use and prompts represent an important window into how these companies might reduce the friction between identifying a problem and moving their overall work forward aided by AI. Our findings suggest that work design deserves a place alongside access to AI tools in how organizations implement their AI strategies.

Workers use AI differently when they move beyond their roles

Workers use AI differently depending on whether the task fits into their role in a particular domain. Workers prompt AI differently for tasks inside and outside their usual roles. When asking about tasks outside their occupation, they write shorter prompts on average than they do for tasks within it. They are also less likely to ask for explanations, how-to guidance, a specific response format, or advice.

At the same time, workers are more likely to provide examples or background when asking AI for help outside their occupations. They are also more likely to ask AI to check or verify something.

One interpretation is that workers are using AI to borrow expertise. Rather than asking AI to teach them an entirely new field, workers may bring a problem and relevant context—such as a document, example, or information from a colleague—and ask AI to help apply knowledge associated with another field. 

Cross-occupation tasks can become recurring parts of new AI workflows

The next question is whether workers return to these activities. Among roughly 6,200 workers observed consistently from April through July, previously used cross-occupation tasks increased from 13.1% of occupation-specific AI activity in April to 25.9% in July1. That pattern is consistent with workers incorporating some cross-occupation assistance into ongoing workflows rather than simply experimenting once. 

A separate analysis puts recurrence in context. Among sampled matched one-month follow-up observations, workers returned to a cross-occupation task used in the previous month 23.6% of the time, compared with 8.4% use of the same task among comparable workers who had no observed use of it in the previous month. Similar gaps appear for within-occupation and general tasks.

Some new activities stick more than others

Recurrence varies considerably depending on the type of work. Workers returned the following month to some cross-occupation tasks at relatively high rates:

• Discussing goods or services with customers: 54%

• Advertising or promotional writing: 44%

• Creating marketing materials: 37%

By comparison, workers returned to the task of explaining financial information about 15% of the time. The average next-month return rate across cross-occupation tasks was 18.5%. These differences may reflect where AI fits naturally into recurring workflows. They could also reflect differences in workplace norms, caution, or the perceived consequences of getting something wrong.