OpenAI Research: How AI is Expanding Occupational Task Crossover
OpenAI Research: How AI is Expanding Occupational Task Crossover
OpenAI research indicates that AI is fundamentally changing the division of labor by enabling "task crossover," where workers perform tasks historically associated with other occupations. An analysis of over 800,000 messages from U.S. ChatGPT users shows that 16.8% of all work-related messages and 43.5% of occupation-specific messages involve tasks outside the user's primary occupation.
The Phenomenon of Task Crossover
Task crossover occurs when work traditionally associated with one occupation appears in the AI usage patterns of people in another. This shift suggests that AI allows employees to handle needs independently rather than relying on a handoff to another specialist.
To isolate this trend, OpenAI categorized messages into "generic" tasks (such as summarizing, scheduling, and writing) and "occupation-specific" tasks. Among non-generic messages, 43.5% fall outside the user's occupation. This trend is most pronounced in the following roles:
- Customer Experience workers: 77% of occupation-specific messages are outside-occupation tasks.
- Designers: 75% of occupation-specific messages are outside-occupation tasks.
- Human Resources workers: 69% of occupation-specific messages are outside-occupation tasks.
- Legal workers: 56% of occupation-specific messages are outside-occupation tasks.
- Marketers: 53% of occupation-specific messages are outside-occupation tasks.
Patterns of Task Distribution
Task crossover does not affect all occupations equally; some roles act as "sources" of tasks that others adopt, while others act as "absorbers" of tasks from other fields.
Engineering and Marketing as Task Sources
Engineering and marketing tasks are the most frequent "travelers" across occupations. Engineering tasks account for 7.4% of messages among workers in other occupations, while marketing tasks account for 8.9%—the highest outward share in the sample. Common examples include technology troubleshooting and working with technical systems.
Design as a Task Absorber
In contrast, design is characterized by high inward crossover. Approximately 35.2% of messages from designers involve work associated with other occupations, but design tasks themselves account for only 1.7% of messages from workers in other fields.
Universal Tasks
Certain tasks have become nearly universal across the sample. Financial calculation and technology troubleshooting are among the three most common outside tasks for all seven occupation groups analyzed.
Impact of Organization Size on AI Usage
The degree of task crossover varies based on the size of the business, with smaller organizations showing a higher tendency for workers to use AI as a generalist tool.
Among average users, the share of outside-occupation tasks decreases as organization size increases:
- Workspaces with 2–5 seats: 18.9% outside-occupation task share.
- Workspaces with over 100 seats: 16.3% outside-occupation task share.
OpenAI suggests that in smaller organizations, where specialist resources are scarcer, the worker closest to the problem is more likely to use AI to solve it rather than delegate it to another function.
Implications for Occupational Change
Because AI usage data provides a real-time window into how work is shifting, it serves as an early signal of occupational change. These patterns of task crossover emerge before they are reflected in formal job descriptions or official labor-market statistics, suggesting that many jobs are likely to reorganize as their day-to-day tasks change substantially.