Technology & Intelligence · March 2026
O*NET AI Exposure Explorer
An interactive tool for evaluating AI exposure at the task level — pick an occupation, adjust exposure scores, and watch weighted impact recalculate live.
O*NET is the U.S. Department of Labor's canonical occupational taxonomy — over 1,000 occupations, each decomposed into discrete tasks with importance and frequency ratings. When Eloundou et al. published "GPTs are GPTs" in 2023, they scored every O*NET task for LLM exposure. This tool lets you do the same thing at the occupation level: select a role, see its tasks, adjust AI exposure per task, and watch the importance-weighted score update instantly. Twenty-five occupations across eight PE-relevant sectors, drawn from O*NET 28.1.
AI Exposure Explorer
24 PE-relevant occupations · O*NET 28.1
Overall Exposure
Financial Analysts · 10 tasks
Analyze financial information to produce forecasts of business, industry, or economic conditions for use in making investment decisions.
LLM can draft narrative analysis; human validates assumptions and model inputs.
Prepare plans of action for investment based on financial analyses.
Requires judgment on risk tolerance and portfolio context; LLM assists with structuring.
Interpret data on price, yield, stability, future investment-risk trends, economic influences, and other factors affecting investment programs.
Pattern recognition assistable by AI; interpretation requires domain context.
Recommend investments and investment timing to companies, investment firm staff, or the public.
Fiduciary judgment and client relationship context are hard to automate.
Evaluate and compare the relative quality of various securities in a given industry.
LLM can summarize filings and compute ratios; comparative judgment is human.
Present oral or written reports on general economic trends, individual corporations, and entire industries.
Report drafting and slide narrative generation are strong LLM tasks.
Gather financial data from internal and external databases and reporting systems.
Structured data retrieval partially automatable; API integration helps.
Determine the prices at which securities should be syndicated and offered to the public.
Pricing requires market timing judgment and regulatory awareness.
Monitor fundamental economic, industrial, and corporate developments by analyzing information from financial publications, investment banking firms, and other sources.
News monitoring and summarization is a core LLM strength.
Collaborate with investment bankers to attract new corporate clients.
Relationship-building and negotiation are inherently human.
Source: O*NET 28.1 · Default exposure scores seeded using Eloundou et al. (2023) rubric · Formula: Σ(exposure × importance) / Σ(importance)
Methodology
Each task's exposure score (0–1) estimates the share of task time that current AI systems (LLMs, code generators, document processors) could meaningfully reduce with access to the task. The overall occupation score weights each task by its O*NET importance rating:
Default scores are seeded using the Eloundou rubric: a task is "exposed" if an LLM or LLM-powered tool could reduce the time needed by at least 50% for a worker at median skill. Scores between 0.1 and 0.4 indicate partial assistance; 0.5–0.7 indicates substantial augmentation; above 0.7 suggests near-full automation potential.
Importance (1–5) reflects how critical a task is to the occupation. Frequency (1–5) indicates how often the task is performed. The explorer weights by importance, not frequency — a rare but critical task matters more than a frequent but peripheral one.
Sources
O*NET 28.1 Database
U.S. Department of Labor / Employment and Training Administration. O*NET OnLine, Version 28.1. National Center for O*NET Development. Occupations, tasks, importance and frequency ratings.
Eloundou et al. (2023)
Eloundou, T., Manning, S., Mishkin, P., & Rock, D. (2023). "GPTs are GPTs: An early look at the labor market impact potential of large language models." Science, 384(6702). Exposure rubric and methodology for scoring AI impact at the task level.