AI Models Disagree on Which Jobs AI Will Destroy: Study Reveals Shocking Inconsistencies

Economists asked ChatGPT, Gemini, and Claude which jobs are most exposed to AI. The result? The models often gave different answers, highlighting a critical flaw in using AI to predict its own impact on the job market.

The Problem with AI Job Predictions

The numbers researchers use to forecast which jobs could be lost to artificial intelligence may be unreliable, and one culprit could be AI itself. Worries about how AI will affect the job market have become one of the most pressing economic questions of our time, and understanding the limits of any forecasts is crucial.

Policymakers want to know which workers could be displaced and need support. Schools, students, and parents want to know which careers might be “AI-proof.” But if the AI models themselves can't agree, how can we trust the predictions?

Key Takeaways

• AI models are inconsistent when predicting job exposure to AI, raising doubts about the reliability of such forecasts.
• The study underscores the need for caution when using AI to inform policy decisions about workforce displacement.
• Understanding the limitations of AI predictions is crucial for workers, educators, and policymakers alike.