OpenAI says more workers are using ChatGPT to do other people's jobs
OpenAI analyzed over 800,000 work-related ChatGPT messages and found that 43.5% of job-specific queries involved tasks outside the user’s profession, a phenomenon termed “task crossover.” Marketing and engineering roles exhibited the highest rates of task crossover, with users performing specialized tasks such as contract reviews, data analysis, and website troubleshooting without formal training in those areas. The effect is more pronounced at smaller companies where dedicated specialist teams
Analysis
TL;DR
- OpenAI analyzed over 800,000 work-related ChatGPT messages and found that 43.5% of job-specific queries involved tasks outside the user’s profession, a phenomenon termed “task crossover.”
- Marketing and engineering roles exhibited the highest rates of task crossover, with users performing specialized tasks such as contract reviews, data analysis, and website troubleshooting without formal training in those areas.
- The effect is more pronounced at smaller companies where dedicated specialist teams are less common, suggesting AI may be accelerating role fluidity in resource-constrained environments.
- OpenAI used the U.S. O*NET occupational database to classify tasks while excluding routine activities like writing, summarizing, and scheduling to focus on non-routine, cross-professional work.
- This trend signals an early shift in job profiles before organizational structures or job descriptions adapt, indicating potential long-term impacts on workforce planning and skill development.
Why It Matters
This finding highlights how generative AI tools like ChatGPT are enabling employees to perform tasks traditionally reserved for specialists, potentially reshaping labor dynamics and reducing barriers to entry across professions. For AI practitioners and organizations, it underscores the need to consider how AI-driven task delegation affects team composition, upskilling strategies, and productivity metrics—especially in small businesses lacking specialized staff. Understanding these patterns can inform better tool design, training programs, and policy frameworks around AI adoption in workplaces.
Technical Details
- Data Source: Over 800,000 anonymized work-related ChatGPT messages collected by OpenAI from enterprise users.
- Task Classification: Tasks were mapped using the U.S. Department of Labor’s O*NET database, which categorizes occupations based on standardized activity descriptors.
- Excluded Activities: Routine cognitive tasks such as drafting emails, summarizing documents, and calendar scheduling were filtered out to isolate novel or cross-domain usage.
- Cross-Domain Metric: A query was classified as “cross-professional” if the task described fell outside the typical responsibilities associated with the user’s self-reported occupation.
- Segmentation Analysis: Usage patterns were broken down by company size, revealing stronger crossover effects in firms with fewer than 50 employees compared to larger enterprises.
Industry Insight
Organizations should anticipate increasing demand for hybrid skill sets as employees leverage AI to take on responsibilities beyond their core roles—particularly in marketing, engineering, legal support, and IT operations. Companies investing in AI integration must proactively update job descriptions, redefine performance expectations, and develop internal training frameworks to manage this transition effectively. Additionally, HR departments may need to reassess hiring practices, favoring adaptability and broad competency over narrow specialization, especially in startups and mid-sized firms where agility is critical.
Disclaimer: The above content is generated by AI and is for reference only.