Too often, anti-trafficking agencies and programs are under-resourced, over-taxed, and lack technical analytics support to connect their efforts with the local realities that shape vulnerability. This pilot project sought to answer a simple question: What if we could use machine learning to equip agencies with clear, actionable insights that enable them to act?

In the noise of all potential risk factors, we looked for:

• Community-specific drivers of human trafficking
• Person-centered combinations of risk factors
• Distinct pathways that lead to human trafficking

Through person-centered analysis, we deployed 200 machine learning models to identify locally specific risk factors. This data supported our partnering agency, at:project (J Bar J Youth Services), in discovering new patterns, so they can create targeted, evidence-based prevention strategies that are focused, efficient, and responsive to community needs.

The analysis provided at:project with evidence-based insights to prioritize data collection focused on the most predictive risk factors and develop targeted prevention strategies for different vulnerable populations based on their unique pathways into vulnerability.

The findings also suggest that effective trafficking intervention and prevention strategies should be tailored to specific risk subgroups rather than applying a one-size-fits all approach.

Many youth in this study faced complex, interconnected risk factors. This study suggests that while certain individual factors (like juvenile justice involvement or substance abuse) come into play, importantly, it was the unique combination of factors that pushed youth into high vulnerability of trafficking. Each unique combination offers opportunities for tailored solutions.