Power plant operators and railroad conductors are more likely to be replaced by AI than artists and writers.
Public opinion holds that creative and white-collar jobs are at the greatest risk from automation. This study uses a Reinforcement Learning feasibility index to show that high-stakes technical roles are actually easier for AI to master. These jobs follow rigorous operational rules that machines can simulate and optimize much faster than humans. Creative roles require an interpersonal spark that current AI architectures still cannot replicate. This flips the script on which careers are safe in the coming decade. Blue-collar technical specialists may face a faster wave of displacement than the poets and painters they once pitied.
What Jobs Can AI Learn? Measuring Exposure by Reinforcement Learning
SSRN · 6659746
Which jobs can AI learn to do? We examine this for every occupation in the US economy. Existing indices measure the overlap between AI capabilities and occupational tasks rather than which tasks AI systems can learn to perform, and as a result misclassify occupations where the gap between present capability and learnability is large. Reinforcement learning in post-training, now the dominant paradigm at the frontier, is structured around task completion and maps more directly onto the task-based