Skip to content

kirso/aiworkindex

Repository files navigation

AI Work Index

V7 is the live structural score: a deterministic two-axis model that separates displacement pressure from demand resilience. V7 adds task-concentration-weighted exposure and a demand-persistence proxy, while retaining V6 baseline fields and historical V4/V5/V6 artifacts for auditability.

Live Site | Global Methodology | Methodology | Calculator | Data

Singapore is the reference implementation. The shared structural baseline also powers global and country pages; local layers add demand, wages, policy, and confidence where evidence is strong enough.

Key Numbers

  • 562 Singapore occupations scored across 9 major groups
  • 88 synthetic roles (product manager, data scientist, delivery rider, startup founder...)
  • 21% face high+ AI risk (118 occupations)
  • SGD 35.1B estimated annual wage pool under high+ structural pressure
  • 492 / 562 occupations have weighted task-primitives evidence
  • 4-source exposure ensemble: Felten AIOE + Anthropic observed usage + Eloundou GPT exposure + ILO 2025

How It Works

Deterministic scoring — no LLM in the scoring pipeline:

  1. Exposure - reliability-weighted 4-source ensemble: AIOE (2021), Anthropic Economic Index (2026), Eloundou GPTs-are-GPTs (2024), ILO Refined Index (2025).
  2. Task concentration - Anthropic task penetration matched to O*NET task statements; concentrated exposure raises structural pressure.
  3. Human bottleneck - Pizzinelli theta from O*NET Work Context (judgment, presence, coordination).
  4. Demand resilience - MOM employment/wage trends, vacancy pressure, SOL/JiD demand signals, and demand-persistence proxy.

The same structural spine is intended to power future country adapters. Singapore remains the reference implementation, while /global defines the comparable baseline for new markets.

task_signal = task_effective_coverage x task_exposure_concentration
exposure_v7 = clamp01(exposure x (1 + 0.20 x task_signal))
displacement_pressure = exposure_v7 x (1 - bottleneck)
headline_risk = displacement_pressure x (1 - demand_resilience)

Published as risk bands (Very Low through Very High) with visible confidence, uncertainty intervals, retained V6 baselines, and historical release artifacts.

Validation

  • BLS cross-country: live convergent cross-check against US BLS projections
  • Cluster-level: directional check against Singapore labour-monitor clusters
  • Release pipeline: validated end to end with published artifacts, checksums, claims matrix, and shadow-model governance outputs
  • Methodology page: aiworkindex.com/methodology

Quick Start

git clone https://github.com/kirso/aiworkindex
cd aiworkindex
bun install
bun run build:release-data  # Refresh all release datasets and metadata
bun run scripts/score.ts    # Score all 562 occupations
bun run validate            # Run release and data-contract validation
bun run dev                 # Start dev server
bun run build               # Build the prerendered static site

Data Sources

Source What Year
MOM Singapore 562 SSOC occupations, wages, employment 2024-2025
Felten AIOE AI exposure per SOC (academic index) 2021
Anthropic Economic Index Observed AI usage (HuggingFace, CC-BY) Jan 2026
Eloundou et al. GPT-4 task-level exposure (Science, 2024) 2024
ILO Refined Index ISCO-08 exposure (52K expert data points) May 2025
O*NET Work Context, Job Zones, Task Statements 2020
MOM SOL 2026 Shortage Occupation List Nov 2025
MOM Jobs in Demand In-demand occupation flags Dec 2025
US BLS Employment projections 2024-2034 (convergent cross-check) Aug 2025

Singapore Context

Each occupation page shows:

  • Education level (O*NET Job Zones → Singapore labels)
  • Progressive Wage Model coverage (57 occupations in 9 PWM sectors)
  • Licensed profession flag (53 strict + 23 partial)
  • Foreign worker dependency (73 very high + 33 high + 45 moderate)
  • SkillsFuture career conversion eligibility (154 occupations)
  • Industry footprint + worker profile from official Singapore labour tables
  • Transition infrastructure from Jobs Transformation Maps, CareersFinder, WSQ, and SkillsFuture / WSG programmes

Data Download

Research Library

  • Public registry: aiworkindex.com/research
  • Machine-readable artifact: static/data/research-library.json
  • Live methodology references are now generated from the same canonical research registry used by reports and release governance.

Tech Stack

  • SvelteKit 5 + Svelte 5 runes (static site, adapter-static)
  • Tailwind CSS v4 + shadcn-svelte (Bits UI)
  • D3.js for visualization layout
  • TypeScript scoring pipeline (Bun runtime)
  • Satori + Resvg for OG image generation
  • Deployed on Cloudflare Workers

Limitations

  • Exposure data age: AIOE is from 2021 (pre-GPT-4). Ensemble with newer sources mitigates but doesn't eliminate.
  • Employment granularity: detailed Singapore occupation counts are not publicly released. estimated_sg_employment_thousands is a labeled sub-major allocation, and wage-pool analysis uses a separate labeled BLS-weighted proxy.
  • Demand-resilience weights: market momentum, vacancy, scarcity, demand-signal, and demand-persistence weights are calibrated, not empirically derived.
  • Cluster-level labour data: Same vacancy/hiring data for all occupations in each of 3 clusters.
  • Synthetic role weights: Expert-assigned SSOC blends, not validated against job posting data.

License

MIT

Author

Kirill So · X

Built with Claude (Anthropic) & GPT (OpenAI)

About

AI Work Index — 562 Singapore occupations and 80 modern roles scored for AI displacement risk. Deterministic three-layer model, no LLM in scoring. Built with SvelteKit 5.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors