ChatGPT data points to work beyond job titles
OpenAI Economic Research says workers are using ChatGPT for tasks that often sit outside the traditional boundaries of their own occupations. In an article published July 27, 2026, the company introduced its Work at the Frontier series and reported an analysis of more than 800,000 messages from U.S. ChatGPT users.The article frames the results as an early signal of changing work, not as a direct measurement of hiring, wages, layoffs or new job titles. OpenAI says the usage patterns may show workers experimenting with new combinations of activities before organizations formally rewrite job descriptions or create new roles.
What OpenAI measured
OpenAI calls the pattern task crossover: work historically associated with one occupation appearing in the AI use of people in another occupation. The research separates broad, generic activities from more occupation-specific work. Writing, summarizing and scheduling are excluded from the occupation-specific crossover measure because OpenAI says those activities occur widely across jobs.On that basis, OpenAI reports that 16.8% of work-related messages and 43.5% of occupation-specific messages concern tasks associated with another occupation. The second figure is the more focused measure because it looks at non-generic messages and asks whether the task belongs inside or outside the user’s occupation.
Task crossover varies sharply by occupation
Among non-generic messages, OpenAI says 43.5% fall outside the user’s occupation. That overall figure masks large differences across occupational groups. The article says outside-occupation tasks account for 77% of occupation-specific messages from customer experience workers, 75% from designers, 69% from human resources workers, 56% from legal workers and 53% from marketers.Those numbers suggest that ChatGPT use may be especially broad in functions where workers frequently need materials, calculations, explanations or process support from neighboring specialties. The source does not say that workers have changed occupations, nor does it say that organizations have formally merged roles. It reports what appears in user messages.
This distinction matters for interpreting the data. A designer asking for help with financial calculation, or a human resources worker asking for technology troubleshooting, is not the same as a permanent job redesign. But repeated use of AI for such tasks may indicate where employees are testing the practical edges of their roles.
Marketing and engineering tasks travel widely
OpenAI says marketing and engineering tasks travel particularly far across occupations. In the analysis, financial calculation and technology troubleshooting each appear among the three most common outside tasks in all seven other occupation groups. Creating marketing materials appears across five other groups and is especially prominent among design users.The article also reports two directions of crossover: how much a group reaches outward, and how often that group’s tasks appear in other occupations. For designers, 35.2% of messages involve work usually associated with another occupation, while design tasks account for 1.7% of messages from workers in other fields.
Engineering and marketing show different profiles. For engineering, OpenAI says 18.5% of messages involve tasks from other fields, while engineering tasks account for 7.4% of messages among workers in other occupations. Marketers devote 24.3% of their messages to tasks associated with other occupations, while marketing tasks account for 8.9% of messages among workers in other fields, the highest outward share reported in the sample.
Business size is linked to different patterns
OpenAI also reports a relationship between business size and crossover among average users. The outside-occupation task share falls from 18.9% for users in workspaces with 2 to 5 seats to 16.3% for users in workspaces with more than 100 seats. The company says this monotonic pattern is not observed among the heaviest users.The article offers a possible explanation rather than a firm causal conclusion. OpenAI says moderate users in smaller organizations may turn to AI when a task would otherwise require another function, while heavy users may have more stable AI-supported workflows. The source does not claim that company size alone causes the differences.
Why the findings are useful, and limited
The findings are useful because they focus on task content rather than occupational labels alone. Labor-market change can be difficult to see if analysts look only at job titles, which may remain stable while day-to-day activities shift. OpenAI’s message-level view is intended to capture some of that earlier movement.At the same time, the evidence is bounded by what the source describes: U.S. ChatGPT user messages in the analyzed sample. The article does not establish economy-wide outcomes, and it does not show whether AI-assisted task crossover improves productivity, changes pay or changes staffing decisions.
OpenAI says the Work at the Frontier series will provide regular data-driven insights intended to inform policy and practice. For now, the most defensible conclusion is narrower: in OpenAI’s data, many workers are using ChatGPT to attempt tasks that historically belonged to other occupational groups.
Sources
Editorial Team - CoinBotLab