We’re pioneering new methods to make AI a tool for empowerment hand-in-hand with groups monitoring the news for reports of human rights violations. Our initiative is developing AI techniques grounded in justice and community needs, prioritizing inclusivity, transparency, and fairness. We partner with a network of groups that monitor news online for reports of human rights or civil rights violations, building databases to support activism, policy change, policy enforcement, and popular protest.
The issues they track include femicide, mass shootings, criminalization of pregnancy, anti-trans violence, and more. Our tool automatically scour millions of online news articles each day to filter them by geography, keywords, and custom-trained machine learning classifiers to narrow in on stories that might be the incidents they track.
These story top candidates are delivered to groups via email and a web-based dashboard. This includes clustering stories by events, and we’re experimenting with automated summarization, subject and date extraction, and more tools to help reduce the time and emotional burden of tracking violent events.
By creating tools that allow communities to harness the power of AI for their own needs, we aim to help reimagine innovative computational systems to amplify diverse perspectives and advance meaningful, community-driven social change. This work is done hand-in-hand with the Data Against Feminicide collective and the MIT Data + Feminism Lab.
