New research from OpenAI reveals that nearly half of all occupation-specific AI interactions involve tasks outside an employee’s formal role, signaling a fundamental shift in labor dynamics.
The traditional division of labor, a cornerstone of economic structure since the industrial era, is undergoing a rapid transformation as artificial intelligence enables workers to bypass formal role boundaries. A study released on July 27, 2026, by OpenAI, titled Work at the Frontier: How AI is Expanding What People Do at Work, provides empirical evidence that nearly half of all occupation-specific AI requests involve tasks traditionally assigned to different professional domains.
Ronnie Chatterji, chief economist at OpenAI, suggests that the integration of large language models is fostering a new level of organizational flexibility. The research, which analyzed over 800,000 ChatGPT messages from U.S. business users, identifies a phenomenon termed task crossover, where employees leverage AI to perform functions outside their core job descriptions.
To isolate meaningful trends, the research team filtered out generic tasks such as email drafting and document summarization, which accounted for 61.5% of total work-related messages. The remaining 38.5% of requests were classified as occupation-specific, mapped against the U.S. Department of Labor’s O*NET database to determine if the requested activity aligned with the user’s formal role.
Within this subset, 43.5% of requests demonstrated task crossover, indicating that workers are increasingly assuming responsibilities previously siloed within other departments. Customer experience professionals exhibited the highest crossover rate at 77%, followed closely by designers at 75% and HR personnel at 69%.
The study also highlights distinct patterns of directionality, noting that some roles act as net importers of tasks while others remain largely insular. Designers frequently borrow from external disciplines but rarely see their own specialized tasks requested by others, whereas engineering tasks are widely sought after by employees across various departments.
Marketing professionals occupy a unique position in this landscape, demonstrating high levels of both task borrowing and task contribution. This dual-directional flow suggests that marketing has become a central hub for cross-functional activity in the age of AI-assisted workflows.
Company size serves as a critical variable in the frequency of these crossovers, with smaller organizations showing higher rates of role-blurring than their larger counterparts. In firms with fewer than 100 employees, AI often functions as an on-demand specialist, allowing staff to bridge expertise gaps without the immediate need for additional hires or external contractors.
The findings offer a leading indicator of organizational change that traditional labor statistics, such as wage surveys and employment counts, have yet to capture. While these metrics remain lagging signals of settled structural shifts, chat-log analysis reveals the fluid, real-time experimentation currently occurring within the modern workforce.
Despite the clarity of the data, the report acknowledges significant limitations regarding long-term economic outcomes. Researchers cannot yet determine if these AI-assisted crossovers result in output quality equivalent to that of a dedicated specialist or if they will eventually lead to a net reduction in specialized headcount.
The broader economic context remains complex, as independent research from the Peterson Institute for International Economics suggests that the study of AI’s impact on labor is still in its early stages. While Goldman Sachs estimates that AI is contributing to net job displacement in specific sectors like data entry and legal support, the current evidence points toward a reorganization of tasks rather than wholesale role elimination.
The primary challenge for policymakers and business leaders will be to distinguish between temporary productivity gains and permanent shifts in labor demand. As organizations continue to integrate these tools, the distinction between generalist and specialist roles may continue to dissolve, necessitating a reevaluation of traditional workforce planning and skill development strategies.
Future research will likely focus on whether this trend toward task crossover correlates with measurable changes in hiring patterns or wage structures. For now, the data confirms that the boundaries of professional identity are becoming increasingly porous in the presence of generative AI.
At Global Business & Economics Journal, we contribute to the future of business and technology with news, innovative research, expert insights, and thought-provoking analysis.
