About This Dashboard
This dashboard brings together yearly employment statistics from Statistics Sweden (SCB) and AI-exposure scores from the DAIOE framework to support research into how AI may be reshaping labour market outcomes across Swedish occupations.
Data Sources
| Source | Description |
|---|---|
| Swedish Occupational Register, SCB | Yearly employment counts and year-over-year changes by occupation, gender, and age group |
| DAIOE Framework | Data-driven AI Occupational Exposure scores across multiple AI capability sub-domains |
Coverage
- Geography: Sweden (national totals)
- Occupation levels: SSYK 2012, all four levels (SSYK1 major groups through SSYK4 detailed units)
- Time range: 2014 to 2024, updated annually
- Employment unit: absolute headcount (e.g. 150,000 = 150,000 people)
Key Concepts
SSYK 2012 The Swedish Standard Classification of Occupations (2012 edition). Organises all occupations into four hierarchical levels:
| Level | Description | Example count |
|---|---|---|
| SSYK1 | Major groups (1-digit) | 9 categories |
| SSYK2 | Minor groups (2-digit) | ~30 categories |
| SSYK3 | Unit groups (3-digit) | ~100 categories |
| SSYK4 | Detailed occupational units (4-digit) | ~400 categories |
DAIOE: AI Exposure Scores Data-driven AI Occupational Exposure scores quantify how strongly the tasks within an occupation may be affected by different AI capabilities. Scores are computed across multiple sub-domains (e.g. language, vision, reasoning) and aggregated as weighted averages at the occupation level.
Percentile Rank Shows where an occupation sits relative to all others on a given sub-domain. A percentile rank of 80 means the occupation scores higher than 80% of all occupations; it is a relative, not absolute, measure.
Exposure Level An ordinal scale from 1 (Very Low) to 5 (Very High) summarising the weighted-average AI exposure score for a sub-domain. Used for quick comparisons; the underlying index score provides more precision.
Employment Change Year-over-year or multi-year percentage change computed from absolute employment counts. Positive values indicate growth; negative values indicate decline. Changes are computed from aggregated employment counts and absolute changes, not by averaging gender- or age-group-specific percentage rates.
Age Groups Employment is broken down by seven age bands: Early Career 1 (16-24), Early Career 2 (25-29), Developing (30-34), Mid-Career 1 (35-39), Mid-Career 1 (40-44), Mid-Career 2 (45-49), and Senior (50+).
Caveats
- AI exposure measures potential task-level exposure to AI capabilities. It is not a prediction of employment decline, job loss, or automation outcomes.
- Year-over-year employment changes may reflect economic cycles, policy changes, survey revisions, or occupational reclassifications unrelated to AI adoption.
- Percentile ranks are relative to other occupations in the dataset. A high rank does not imply a high absolute exposure score, and rankings may shift as new occupations or years are added.
- At SSYK4, many detailed occupational units have small employment counts; year-over-year changes may be volatile.
- The 3-year and 5-year change figures compare the current year against three or five years prior. They will be null for occupations with insufficient history.
About the Project
This tool is developed by the AI-Econ Lab as part of ongoing research into the intersection of artificial intelligence and labour markets. For questions or collaboration enquiries, please visit ai-econlab.com.