Context: The Importance of MMIP Data

In 2021, UIC initiated the creation of its MMIP policy tracker, which tracks “all state-wide or national legislation relating to MMIWGT2S in the United States.” The goal of this tracker is multi-faceted: to demonstrate the current state of MMIP across the country, to monitor progress toward supporting survivors and mitigating the crisis, and to make it clear which jurisdictions are neglecting the issue.

Indigenous communities face a legacy of data genocide—intentional erasure through a lack of data and insufficient funding for new data collection. This intentional erasure has long contributed to a lack of visibility for Indigenous issues, particularly in addressing the MMIP crisis. UIC needed a more efficient system to manage and analyze MMIP-related data, one that could both improve our data collection processes and empower our advocacy efforts.

By leveraging ChatGPT and automating legislative data analysis, we are addressing this data genocide head-on, giving Indigenous communities a way to reclaim and generate culturally relevant data at scale. This project is not just a technical solution; it is a strategic tool in the fight for Indigenous representation and visibility in the legislative space.

Challenge

The initial challenge was standardizing the data collection and storage process. As a result of a previously unstandardized and unclear process, UIC had conflicting datasets that needed to be merged without losing any critical information.

For example, the “Sponsors” and “Indigenous Sponsors” fields had no systematic way of identifying which sponsors were Indigenous. This often led to underreporting and inaccurate data. There was no centralized or automated way to identify Indigenous sponsors, which meant team members had to manually research or rely on name recognition to identify Indigenous legislators, making the process inefficient and prone to error.

Additionally, mixing qualitative and quantitative data blocked us from providing automated reports and metrics, requiring manual updates of over 25 columns in the spreadsheet.

Solution / Process

To address these challenges, we began by merging the conflicting datasets into a unified Excel document. Afterward, we transitioned to AirTable, enabling better standardization of data formats, collection, and automated visualizations through its interface builder. I also created a database of Indigenous politicians and legislators to automate the identification process of Indigenous sponsors. This tool, built by scraping data from Wikipedia (the only public source available), allows us to accurately and efficiently identify Indigenous sponsors in legislation, removing the need for manual research.

The Indigenous sponsor identification database is available as a standalone tool, as well as being integrated into the AI policy analyzer. The analyzer, developed in Python, pulls legislative details from the Legiscan API and uses ChatGPT to parse legislative texts through culturally tailored questions, generating new data for analysis. This tool allows us to automate data generation, compressing what was once an hour-long process into mere minutes, significantly boosting productivity.

The AI-generated policy analysis is then reviewed by volunteers and the MMIP Program Associate before being uploaded into our AirTable database. This process balances automation with human oversight to maintain accuracy and ensure cultural sensitivity in data handling.

Results

The introduction of ChatGPT and the AI policy analyzer, along with the transition to AirTable, has transformed UIC’s data collection and reporting processes. We’ve significantly reduced the time required to process legislative data, turning a manual task of over an hour into a process that takes just minutes. This automation helps Indigenous communities push back against the ongoing data genocide, ensuring that we can efficiently gather and analyze data in real-time to support advocacy efforts.

Learnings

A key lesson from this project has been understanding the balance between automation and human oversight when dealing with culturally sensitive data. Clear data governance protocols were critical as we expanded the team, and we’ve learned the importance of establishing workflows that enable volunteers to contribute effectively while ensuring the accuracy and integrity of the data.