Multidimensional Scam Risk Index

A decision-support tool that scores the structural conditions that let scam compounds, and the forced criminality inside them, take root across Southeast Asia.

Coverage starts with Southeast Asia, where the compound model took hold. UNODC reported in 2025 that as crackdowns intensify, Asian-led scam centers have been found operating in Africa, South Asia, the Middle East, and Pacific islands. (UNODC) The framework is designed to extend to new regions once sourced data and validation exist.

In development · Location scores for vetted partners only

The problem

Scam compounds in Myanmar, Cambodia, Laos, and elsewhere hold people who were trafficked and forced to run online fraud. The UN human rights office estimated in 2023 that at least 120,000 people in Myanmar and 100,000 in Cambodia may be held in these operations. (OHCHR)

The organizations positioned to respond each see only part of the picture: NGOs see survivors, banks see money flows, and policy teams see governance. MSRI combines those views into one consistent, sourced assessment of where the enabling conditions are strongest.

How it works

Each location is scored from 0 to 100 on five dimensions, which combine into a weighted composite. Every score carries its sources, an as-of date, and an uncertainty band based on the confidence of the evidence behind it.

Governance & rule of law25%

Labor & migration25%

Financial flows20%

Digital platform15%

Protection capacity15%

Scores are diagnostic, not predictive. They describe the structural susceptibility of places. They never attribute anything to people or communities, and they don't predict specific incidents. Stronger survivor protection lowers a location's score.

Read the methodology

Who it's for

NGOs and survivor services

Where to pre-position referral contacts, and where enforcement is likely to criminalize victims.

Financial institutions and AML teams

Which geographies carry the laundering infrastructure behind scam proceeds.

Policy teams

Which governance and protection gaps sustain compounds, and where crackdowns displace them.

Status

MSRI started as a graduate capstone at George Washington University's Elliott School of International Affairs (2026). The prototype used illustrative data. It's now being rebuilt on real, cited sources.

As of October 2026, 11 locations in six countries have been researched. Six have enough evidence to score. Five are held as insufficient rather than estimated. Location scores are shared only with signed-in partners, never on public pages.

Indicator framework: 5 dimensions, 14 indicators (done)
Scoring engine with uncertainty bands and low-confidence flags (done)
Source feasibility review for every indicator (done)
Source registry and license review (done)
Country baselines: governance, corruption, income, TIP tiers, FATF lists (done)
Tiered access for vetted partners (done)
Evidence database with two-person review (done)
Automated evidence feeds: ACLED conflict events, US and UK sanctions (done)
Risk map and location briefings for partners, scored from cited evidence (done)
Weighting explorer (done)
Incident timeline that separates claimed, verified, and disputed enforcement (done)
Financial flows data from bank or financial intelligence unit partners (planned)
Extend coverage beyond Southeast Asia (validate indicators for new regions) (planned)

Country context Where the data comes from