No longer flying blind: How collective intelligence can help prevent forced labour

Forced labour remains one of the most stubborn human rights challenges we face. Nearly 28 million people are still trapped in coercive work across industries, across borders, and deep inside global supply chains.

This is not because the world has been idle. Governments have passed laws. Companies have invested heavily in due diligence. Civil society, often with far fewer resources, has pushed relentlessly for change. Commitment and visibility has been increasing. The tools are more sophisticated than ever. And yet, progress has struggled to keep pace with the scale and complexity of the problem. In some areas, real progress has been made, but those gains remain uneven and hard to sustain over time.

What is missing today is not will alone, but the system level capability to connect effort in ways that actually lead to impact. Even the most serious actors are often forced to make decisions without timely, trusted or usable intelligence. Too much of the work depends on partial information and delayed signals. Good intentions matter, but on their own, they do not deliver change.

The global response remains fragmented. Governments, businesses, international organizations, civil society organizations and worker groups each hold important pieces of the puzzle, but no one consistently sees the whole picture. We are surrounded by data, yet still struggle to turn it into insight that leads to action.

Over time, this fragmentation can reinforce a vicious cycle. Limited visibility weakens accountability. As accountability weakens, trust begins to erode. And once trust starts to fray, coordination becomes increasingly difficult and fragile. Without coordination, even sustained and serious effort falls short of producing system wide change.

Breaking this cycle requires a shift in how we understand the problem itself. Forced labour is not a series of isolated compliance failures – it is a systemic challenge that calls for a systemic response. The Global Data Partnership Against Forced Labour was created with that understanding at its core, designed to pull together a set of reinforcing levers that make collective impact possible.

As described in this recent white paper, the Partnership starts from a simple but often overlooked insight: forced labour persists not because data is missing, but because it remains disconnected. Governments collect inspection and migration data. Companies manage audits, supplier records and conduct due diligence. Civil society and worker organizations document grievances and lived experience. Each of these efforts is valuable on its own. Too rarely do these insights come together in ways that support prevention, accountability or coordinated action.

As a result, many actors are effectively flying blind. Decisions are made based on incomplete signals, while exploitation continues unseen within complex recruitment pathways and opaque supply chains.

What the Partnership offers is a trusted, precompetitive way to connect these insights securely, without asking anyone to give up control of their data. By enabling shared intelligence while respecting sovereignty, privacy and consent, it aims to reverse the dynamics that have allowed forced labour to persist for so long.

Five systemic levers sit at the heart of this approach:

1. Visibility: Making risk visible at system level

Forced labour flourishes where visibility is weakest. Warning signs often exist, but they are scattered across disconnected systems. Worker grievances, recruitment fee data, inspection findings and audit results each tell part of the story, yet too rarely inform one another.

In practice, this makes it possible to see across systems, not just within individual silos. Using a federated approach, insights are analyzed where data already sits and combined at the level of intelligence rather than raw information. This allows patterns of risk to emerge that no single dataset could reveal on its own.

Worker voice is central to this effort. Workers are often the first to experience and report exploitation, yet their perspectives are among the least integrated into formal systems by design. When worker-generated data is included safely and responsibly, visibility becomes broader, earlier and more firmly grounded in lived reality.

Prevention depends not just on seeing risk, but on seeing it in time to act. Without timely visibility, intervention arrives too late, after harm has already been done.

2. Accountability: From isolated responsibility to shared accountability

When evidence is fragmented, responsibility fragments with it. Governments enforce labour standards without insight into supply chain realities. Businesses carry out due diligence without visibility into recruitment practices or enforcement trends. Civil society pushes for remedy without access to wider system signals. Investors look for credible indicators without a coherent evidence base.

The result is accountability that largely remains vertical and confined within institutional boundaries.

Shared intelligence changes this dynamic. When actors work from a common, trusted understanding of risk and response, accountability becomes clearer and more actionable. Responsibilities can be aligned across institutions, and actions can reinforce one another rather than pulling in different directions.

This shift strengthens impact, but it also improves effectiveness. Shared intelligence reduces duplication, strengthens credibility and helps organizations meet rising expectations more efficiently.

3. Coordination: Turning parallel efforts into collective action

Across sectors, the commitment to addressing forced labour is real. Governments investigate. Companies audit. Civil society documents harm. Yet too often these efforts move forward in parallel, leading to duplication, audit fatigue and persistent blind spots.

Coordination often falters not because collaboration lacks value, but because it often feels risky or misaligned with existing incentives. Sharing insight can seem costly, while acting alone feels safer and more familiar.

At its core, the Partnership treats data as a shared asset that creates collective advantage. By connecting existing systems rather than replacing them, it allows initiatives to build on one another. Coordination becomes safer when it is designed into the infrastructure, supported by governance and grounded in mutual benefit.

As coordination improves, effort shifts away from repetitive assessments and toward targeted, preventive action.

4. Trust: By design, not by assumption

Data related to forced labour is sensitive, complex and often deeply personal. Without strong safeguards, caution is not only understandable, it is essential.

Trust within the Partnership is built deliberately. Privacy preserving approaches ensure that insights move, not raw data. Governance frameworks set clear expectations for how information can be used. Ethical standards ensure data supports prevention, remedy and accountability, and nothing beyond that.

Trust also depends on confidence in quality. Shared intelligence must be reliable, verifiable and transparent in how conclusions are reached. When actors trust both the safety and the integrity of the insights they receive, participation becomes possible and action more likely.

Trust is not the starting point. It is the outcome of systems designed to earn it.

5. Leadership: Choosing collective impact over isolated compliance

Data and tools can illuminate risk, but leadership is what turns insight into action. Ending forced labour requires leaders who are willing to align around shared evidence, accept shared responsibility and prioritise prevention alongside compliance, rather than relying on reactive approaches alone.

This is not about individual heroism. It is about institutional leadership. Leadership that recognizes no single actor can solve forced labour alone. Leadership that chooses collaboration over isolation, transparency over selective disclosure and long-term solutions over short term fixes.

By participating in shared systems, leaders strengthen not only collective outcomes, but also their own ability to meet growing expectations around due diligence, transparency and ethical governance.

The Global Data Partnership Against Forced Labour brings together powerful levers into a coherent model for systemic change. Through federated data collaboration and shared governance, we create a virtuous cycle of visibility, accountability, coordination, trust, and leadership.

This is no longer theoretical. The same data-driven, AI-powered tools that enterprises use to boost performance and efficiency—through advanced analytics, automation, insights, and traceability—are now being applied to close the information gaps that hinder the fight against forced labour. These capabilities already help organizations manage risk and complexity across industries. The Partnership channels them deliberately and responsibly toward one of the most entrenched human rights challenges of our time.

A proof of concept in Thailand shows this approach in action. Using multilingual natural language processing and open-source technologies, it demonstrates how federated data and agentic AI can transform fragmented information into collective intelligence—flagging patterns and risk hotspots. By enabling secure, cross-system analysis while keeping data protected at its source, it proves that safe collaboration is both technically feasible and institutionally possible.

The goal isn’t to create another platform. It’s to build connective tissue—linking existing efforts and shifting from isolated interventions to a system that can predict and prevent forced labour before it happens.

Breaking the cycle of forced labour requires collective action grounded in shared intelligence. Governments, businesses, international organizations, civil society and worker groups all have a role to play, not in isolation, but together.

By contributing insights responsibly and acting on shared evidence, participants can help build a system where forced labour becomes a preventable risk rather than an enduring reality.

The question is no longer whether this is possible. It is whether leaders will choose to act together while the window remains open.