Categories: FAANG

How Palantir AIP Supports Polaris In Its Anti-Human-Trafficking Work

Advancing the Mission While Protecting Victims’ Sensitive Data

Introduction

Human trafficking is a pervasive issue that affects millions globally. At Palantir, we are committed to leveraging our advanced data analytics platforms to support organizations like Polaris, who are at the forefront of the fight against human trafficking. This blog post explores how Palantir’s Artificial Intelligence Platform (AIP) supports Polaris in their mission to end sex and labor trafficking and support survivors on their journeys to freedom.

Polaris: A beacon of hope

Polaris, named after the North Star that guided slaves to freedom during the American Civil War, is a beacon of hope for countless victims of human trafficking. Founded in 2002, Polaris has grown to become a leading organization in the anti-trafficking field. Its mission is to end sex and labor trafficking and support survivors on their journeys to freedom. They envision a world where the vulnerable are not exploited by the powerful for profit.

Polaris operates through three main areas of work:

  1. Direct Response: Providing immediate support to victims through the National Human Trafficking Hotline and the Polaris Resilience Fund.
  2. Research and Intelligence: Gathering and analyzing data and sharing actionable insights to understand and combat trafficking.
  3. Campaigns and Public Engagement: Raising awareness, mobilizing the public, and advising policymakers to take action against trafficking.

We primarily support Polaris by providing tools that help enable Polaris’s research and intelligence work, as well as direct response efforts.

The Genesis of the Partnership

Our partnership with Polaris began over a decade ago and has evolved significantly over the years. Polaris CEO Catherine Chen describes the Palantir-Polaris relationship as “invaluable,” noting “the longevity of the partnership has allowed us to build a deep understanding and mutual trust, enabling both organizations to evolve and continuously discover innovative strategies for tackling new challenges.”

At the onset of our partnership, we provided Polaris with software to develop dashboards specifically designed to help the National Human Trafficking Hotline with assistance dispatching, as well as link analysis tooling through our Gotham platform. These tools simplified complex data, facilitating the identification of trafficking hotspots and promoting efficient resource allocation. Additionally, this empowered Polaris to pursue policy changes in a data-driven manner.

As our partnership deepened and the range of our solutions expanded, Polaris transitioned to utilizing our Foundry platform. This platform, designed with adaptability in mind, accommodates Polaris’s evolving needs. Leveraging the power of Foundry’s Ontology and Object Explorer, Polaris has been able to go beyond its traditional link-network analysis. This has empowered Polaris to develop bespoke apps and analytical tools, crafted to meet its specific needs, while further utilizing AIP capabilities to harness the power of real-time data analysis.

Polaris’s Data Vision and Implementation

Polaris’s data vision is to bridge the gap between understanding the problem of human trafficking and taking actionable steps to mitigate it. Polaris holds one of the largest data sets on human trafficking, gathered from its hotline and various research efforts. “This data,” Polaris’s CEO emphasizes, “is crucial for informing policies, guiding interventions, and raising public awareness.”

In pursuing this vision, Polaris operates on two main principles:

  1. Survivor-Centered Approach: Survivors are the experts. Polaris is dedicated to empowering survivors, enabling them to regain control and play an active role in driving the research and decision-making processes.
  2. Collaborative Model: By working closely with allies from various sectors, Polaris leverages diverse talents, resources, and perspectives to solve complex problems.

One of the standout projects that exemplifies this approach is the National Survivor Study (NSS). Led by Sara Woldehanna, Polaris’s Managing Director of Data, Strategy, and Impact, this study was co-designed with survivors to ensure that research addressed the most relevant needs of the community and was implemented in a way that was respectful and empowering of the survivors who participated. The study utilized our Foundry platform to analyze data, making it possible for Polaris to produce actionable insights quickly and efficiently.

The collaboration model for this project included:

  1. Survivor and Anti-Trafficking Partners Leadership: Survivors and anti-trafficking partners were at the forefront, determining the research questions and methodologies.
  2. Industry Partner Engagement: Questions from industry partners, such as financial institutions interested in combating financial abuse, were included.
  3. Technology Integration: Palantir Foundry was critical in building the technology infrastructure, including the survey tool, phone banks, payment methods, and data analysis capabilities.
  4. Widespread Impact: NSS findings have been used in various sectors, from the banking industry to hospitality, and have empowered survivors to advocate for themselves.

To amplify the impact of its data and approach, Polaris recently launched the Data to Impact (DTI) Initiative, a multi-year plan for building capacity to produce and analyze data more efficiently and at a deeper level — and more effectively combat trafficking at scale.

Evolution of the Polaris-Palantir Partnership

The technological foundation for Polaris’s data vision and DTI Initiative relies heavily on its long-term collaboration with Palantir. Over the course of more than a decade, our partnership with Polaris has evolved significantly. Initially, we provided an implementation team to help Polaris navigate complex data landscapes alongside Polaris staff. As its data capabilities grew, Polaris began to take more ownership of the technical workflows, integrating them into its daily operations. Today, Polaris’s teams — from engineers and data scientists to operational users — manage intricate workflows on their own.

In light of Polaris’s DTI Initiative, one of the recent key milestones in the partnership was Polaris’s participation in Palantir’s long-revered Hack Week tradition, in which we set aside day-to-day work to explore novel approaches to solving complex problems. Annabelle Larose, a privacy and civil liberties engineer at Palantir, who has spearheaded our joint project for a number of years, nominated Polaris as a Hack Week partner project. Over several months, Palantir and the Polaris team scoped out projects with a dual objective: to instigate transformative changes in Polaris’s data ecosystem through immediate system modernization, and to proficiently utilize the technologies offered by Palantir AIP.

For Polaris, project feasibility is typically limited by staffing. However, this initiative, which would ordinarily take Polaris 3–4 years to undertake, was condensed into an impressively efficient timeline. This was made possible as a diverse team of Palantir employees, including software engineers, product designers, and data scientists, collaborated with Polaris’s Learning, Innovation and Data Systems (LIDS) team. Within five business days and through intense collaboration with Polaris staff, the Polaris-Palantir Hack Week team focused on establishing three new workflows:

  1. Prototyping a New Customizable User Interface: The Data Quality Assurance team within LIDS was operating on an external platform that lacked the agility and smooth user experience they needed. Users often had to manage multiple sessions and extract information from disparate sources. The proposed user interface aimed to consolidate all information into a single, comprehensive dashboard.
  2. AI-powered Data Extraction: Analysts faced the time-consuming task of distilling case notes. Machine learning models automated data extraction from unstructured text, improving efficiency and incorporating human-in-the-loop principles to ensure accuracy every step of the way.
  3. Real-time Data Visualization: Data analysis is crucial, but downstream tools can further visualize trends and issues within the data. We focused on implementing heat maps, which provided a convenient way to identify patterns at the state, county, and district levels.

For the successful execution of this project, Polaris provided the Hack Week team with access to synthetic (notional) rather than the highly sensitive real data of Polaris. This approach reinforced the recognized privacy principle of data minimization without affecting the quality of the insights obtained. Palantir engaged closely with Polaris and demonstrated significant investment in enabling Polaris to effect tangible modifications. Throughout the process, Polaris remained the decision-maker at every stage.

“Through the Palantir-Polaris Hack Week initiative, we achieved in a week what would have otherwise taken us years to accomplish.”
 — 
Polaris CEO Catherine Chen

The success of this Hack Week effort demonstrated the strength of focused collaboration and established a foundation for future projects. AIP and Foundry have since assisted Polaris in improving workflows for both developers and non-technical users. These enhancements have encompassed various purposes, from utilizing machine learning models to drive extended data analysis projects to expanding the initial project scope. The focus is now on amplifying analytical outcomes while also automating time-consuming tasks.

The Importance of Governance and Data Privacy

Data privacy is a critical concern for Polaris, and they chose to partner with Palantir because of our shared commitment to data protection and privacy-first principles. As with all of our customers, Palantir’s role is to provide software for our customers to use with their data. Polaris closely manages all the data it collects, governed by strict regulations and privacy policies to ensure the protection of all those who engage with them. Polaris takes the responsibility of holding and executing its institutional mission through the use of this highly sensitive information very seriously, as insights derived from these datasets originate from society’s most vulnerable individuals, who must be protected.

In its pursuit of justice for victims of trafficking, Polaris is acutely aware that this is not a battle they can wage alone. Polaris holds itself to the same standards it expects from partners, including software providers like us. Polaris views us as a trustworthy partner due to our extensive experience in providing analytical tools to organizations that handle highly sensitive data while maintaining stringent security standards.

It is critical to note that we do not collect, store, or sell data. Nor do we use our clients’ data to train proprietary AI or machine learning models. Similar to Polaris, we do not facilitate the movement of data between organizations except where those specific organizations have entered into an agreement.

Equally important, Polaris values Palantir’s approach to privacy, which enables organizations to implement privacy frameworks tailored to their needs. This includes the use of encryption, data minimization, and robust access controls, all through easy-to-use privacy and governance-enhancing tools directly within the platform.

Why This Partnership Works

Several factors contribute to the success of the Polaris-Palantir partnership:

  1. Shared Values: Both organizations are committed to responsibly enabling and undertaking the work of understanding complex information and making informed decisions.
  2. Trust and Longevity: A long-term relationship built on trust allows for deeper collaboration and innovation.
  3. Flexibility and Support: We provide the tools and support without constraining customers to particular workflows, giving Polaris the freedom to focus on its mission.
  4. Collaborative Approach: Respecting the expertise and needs of Polaris, while working together to develop solutions tailored to shared goals, is crucial. This approach is far more effective than simply seeking data and running analysis in silos, without collaboration.

Call to Action: The role of technology in supporting NGOs

Technology can be a game-changer for NGOs, providing the tools needed to analyze complex data, make informed decisions, and scale their impact. For technology firms, supporting NGOs is not just about philanthropy; it’s about leveraging their expertise to help solve real-world problems.

For business leaders and technology firms, the Polaris-Palantir partnership offers a blueprint for how to support NGOs effectively. By providing resources, time, and talent, companies can help build the capacity of organizations like Polaris to realize their vision.

The Palantir-Polaris partnership exemplifies how technology can be harnessed to fight human trafficking. Through a blend of shared values, mutual trust, and innovative technology, we are making substantial progress in comprehending these issues. This understanding aids us in addressing these complex challenges, raising awareness and providing support for those most in need. For other companies considering partnerships in the fight against trafficking, the message is clear: support, collaborate, and empower. Together, we can envision a world where the vulnerable are no longer exploited by the powerful.

See Palantir’s Modern Slavery and Human Trafficking Statement.


How Palantir AIP Supports Polaris In Its Anti-Human-Trafficking Work was originally published in Palantir Blog on Medium, where people are continuing the conversation by highlighting and responding to this story.

AI Generated Robotic Content

Recent Posts

Unleash the power of generative AI with Amazon Q Business: How CCoEs can scale cloud governance best practices and drive innovation

This post is co-written with Steven Craig from Hearst.  To maintain their competitive edge, organizations…

21 hours ago

Election Denial Conspiracy Theories Are Exploding on X. This Time They’re Coming From the Left

Conspiracy theories about missing votes—which are not, in fact, missing—and something being “not right” are…

22 hours ago

AI-driven mobile robots team up to tackle chemical synthesis

Researchers have developed AI-driven mobile robots that can carry out chemical synthesis research with extraordinary…

22 hours ago

Aquatic robot’s self-learning optimization enhances underwater object manipulation skills

In recent years, roboticists have introduced robotic systems that can complete missions in various environments,…

22 hours ago

Best AI Tools for Business

Overwhelmed by manual tasks and data overload? Streamline your business and boost revenue with the…

2 days ago

Building a Robust Machine Learning Pipeline: Best Practices and Common Pitfalls

In real life, the machine learning model is not a standalone object that only produces…

2 days ago