NLP Occupational Emergence Analysis: How Occupations Form and Evolve in Real Time -- A Zero-Assumption Method Demonstrated on AI in the US Technology Workforce, 2022-2026

Researcher David Nordfors has published a groundbreaking study on arXiv titled 'NLP Occupational Emergence Analysis' that challenges conventional wisdom about AI careers. Using a novel 'zero-assumption method' that analyzes 8.2 million US resumes from 2022-2026, Nordfors proposes that genuine occupations form as 'co-attractors' where shared vocabulary and practitioner cohesion reinforce each other. The method independently tests vocabulary cohesion and population cohesion without relying on predefined job titles or taxonomies, representing a significant methodological advance in labor market analysis.

Applied specifically to AI, the analysis reveals a striking asymmetry: a cohesive professional vocabulary around terms like 'LLM,' 'fine-tuning,' and 'RAG' formed rapidly in early 2024, but the practitioner population never cohered into a distinct occupational group. Instead of creating a new 'AI Engineer' occupation, the pre-existing AI community dissolved as the tools went mainstream, with expertise being absorbed into existing software engineering, data science, and product management roles. This suggests AI is functioning as a diffusing technology that enhances existing careers rather than creating a new standalone occupation.

The paper raises important questions about whether formalizing an 'AI Engineer' occupational category could catalyze the missing population cohesion around the already-formed vocabulary. This has significant implications for workforce development, educational programs, and corporate hiring strategies. The findings suggest that companies should focus on upskilling existing talent rather than hiring specialized AI roles, and that educational institutions should integrate AI skills across existing curricula rather than creating standalone AI degree programs.