Student Blogs

Discover a collection of student insights through their blogs, where academic journeys and personal stories come to life.

Lattie Thomas at the 2024 Eradicate Hate Summit

The Eradicate Hate Global Summit was formed after the October 2018 massacre at Pittsburgh’s Tree of Life Synagogue. The Summit brings together multi-disciplinary global experts, leaders, and community members who are committed to ending all forms of hate speech and violent extremism. This year, several GSPIA students had the opportunity to volunteer at the Summit and attend discussion sessions. Lattie Thomas is one student who experienced what it was like to be a part of this event.

Read about Lattie's reflections on the Summit here or view below: 

Rising Above Hate: Inspiring Change and Community at the 2024 Eradicate Hate Summit

In October 2024, Pittsburgh hosted the fourth annual Eradicate Hate Global Summit at the David L. Lawrence Convention Center, hoisting itself into the epicenter of a global movement against hate-fueled violence. As speakers and guests began to arrive at 6 a.m., the city was still cloaked in darkness. However inside, a collective sense of hope, urgency, and purpose pulsated through the convention center. 

I had the privilege of volunteering for two days of the three-day summit, working alongside employees from Jones Day Law Firm’s Pittsburgh office. In conversation, they shared that one of their firm's partners, Laura Ellsworth, co-founded the summit with Mark Nordenberg in response to the 2018 attack at the Tree of Life Synagogue. This conference has since grown into a globally recognized event that includes world leaders, policymakers, scholars, students, and activists among its attendees. 

This year’s summit tackled the alarming rise in political violence. The keynote address from Congresswoman Gabby Giffords, a survivor of an assassination attempt, highlighted the importance of this surge of violence, and the urgency of collaborative action. Experts delved into the escalation of hate crimes and explored strategies to combat politically motivated aggression.  

Between volunteering shifts, I attended a session titled "Combining the Grassroots Power of Targeted Communities." Moderated by Ryan Greer, President of Bedrock, the discussion emphasized the importance of uniting different advocacy groups to amplify their collective impact. Bedrock is an organization dedicated to eliminating hate-fueled violence by connecting targeted communities, policymakers, advocacy groups, and other stakeholders to coordinate and strengthen their efforts at a national level. Greer facilitated an insightful conversation with panelists such as Maya Berry, Executive Director of the Arab American Institute, Andy Kang from the Asian American Foundation, and Rucha Kaur from the Sikh Coalition. 

The panelists spoke on the polarizing nature of politics within the United States, and how this polarization can fuel hate and violence. The leaders also shared personal accounts of how their organizations have supported each other, especially in times of crisis. They emphasized that hate is often designed to divide, but collaboration builds a stronger, unified front against discrimination and violence. A win for one community is a win for all. The real enemy is not other marginalized communities—it is the spread of harmful rhetoric and misinformation weaponized to incite violence. 

The Eradicate Hate Global Summit continues to serve as a beacon of hope and a catalyst for real-world solutions. By fostering interdisciplinary collaboration and engaging the next generation of leaders, the summit is reshaping the fight against hate on a global scale. As the event concluded, attendees left with renewed determination and actionable strategies, reinforcing Pittsburgh’s role as a leader in the global movement to combat hate and build a more just, inclusive world.

About the author: Lattie Thomas is a current MPA student concentrating on Governance and International Public Management. She holds a commitment to public service, advocacy, and community engagement. Read more about Lattie in her student bio.

Katie Whipkey on the 2024 UN Summit of the Future 

The Ford Institute supported GSPIA student Katie Whipkey in attending the Summit of the Future in New York City earlier this year.

Read about Katie's experience here or view below:

Hope and Fear: Reflections on the UN Summit of the Future 2024

The environment is contagious. The halls are buzzing with people from around the world, speaking dozens of languages, wearing their best attire, and eagerly discussing the biggest crises of our day. This is the scene at the United Nations (UN) Summit of the Future. Coordinated by the UN, the Summit was formed to gather governmental and civil society high-level leaders together to build trust and coordinate global cooperation in order to achieve agreed upon goals and address emerging threats and opportunities. The Summit was convened because global commitments – namely the Sustainable Development Goals (SDGs) – are not on track to be met, especially in light of the rapidly changing and increasingly complex multilateral governance context. The aim was to reinvigorate an environment of multilateralism and reaffirm existing commitments, including the UN Charter.

The Summit positioned itself as a ‘once-in-a-generation opportunity for change’ to pivot back toward a more collective approach to overcoming current geopolitical realities, global challenges, and emergent technologies to build a better world. The outcome of the Summit was the Pact for the Future, including two critical annexes: a Global Digital Compact and a Declaration on Future Generations. It led up to the 79^th United Nations General Assembly (UNGA79). The hope was the Summit, and the associated Pact, could set the tone for UNGA79 by reaffirming a global commitment to multilateralism, inclusion of civil society and young people in dynamic processes, a more gender-equal world, and a plan to overcome climate inaction.

My experience at the Summit of the Future Action Days echoed the diametric messages heard throughout the sessions and halls: hopefulness and fear. I heard and felt a strong message of hope—we can change, and we must. If we listen to the groups closest to the challenges, together we can come up with equitable and sustainable solutions. Specific solutions raised include adequately financed and localized development programs, nuclear non-proliferation and disarmament, female leadership in all spaces of decision-making, and so many more. In the words of a young feminist activist Ayshka Najib, “To be hopeless is to be privileged.” However, fear was also palpable. Young activists shared that they struggle to see a future for themselves that they want. They admonished the reversal of women’s and queer rights, tokenism of their inclusion, and, of course, the seemingly insurmountable climate crisis.

As someone who is researching civil society engagement in feminist foreign policy (FFP) in international spaces, being part of the Action Days was vital. I had an opportunity to hear directly from actors central to the fight for gender equality, women’s rights, and feminist foreign policy alongside current policy decision-makers. The messages could not be clearer and more consistent: progress toward gender equality is under threat and backsliding. Systemic solutions that are at the heart of FFP – such as gender equity, disarmament, climate justice, corporate accountability, inclusive and participatory decision-making, and decolonization – are essential. While government actors tended to be more guarded in their rhetoric and commitments than the progressive push demanded by civil society, I was moved by several female governmental leaders from countries with FFPs – including Mexico, Spain, and Slovenia – who powerfully supported FFP as a realistic and sustainable solution. Nadine Gosman, President of the National Women’s Institute of Mexico shared, “FFP is a way to influence construction of more just and prosperous societies that center wellbeing of women and girls. A (peaceful) future without women at the center […] is not possible.”

So, what’s next? There is a Pact for the Future in place that outlines an Action Plan. However, it doesn’t offer an implementation plan. That is now in the hands of governments and civil society to take the commitment forward. My own next steps are to follow the process of feminist civil society engagement with FFP governments to track how they can, together, utilize these spaces for influence to promote the principles and approaches offered by FFP to advance the ideas of the Pact and other related international commitments.

About the author: Katie Whipkey is a 2^nd year PhD student at the University of Pittsburgh Graduate School of Public and International Affairs. Her research focuses on feminist foreign policy and the intersection of government and civil society in international arenas. Through the Ford Institute for Human Security, Katie attended the Summit of the Future Action Days and several other UNGA79 events and side events to support her work.

Amanda Zaner on Hate Speech & Social Media Content Moderation

Following her work for the Ford Institute's collaboration with the Center for Democracy & Technology (CDT) on our recent research brief, Amanda wrote a blog post featured on CDTs website.

Read Amanda's Blog here or view below: 

Intersectional Disparities within Automated Hate-speech Detection Across US Centered Social Media Content

Navigating the socio-linguistic nuances of communication across social media platforms is no simple task. When it comes to online content moderation, platforms frequently rely on automated detection systems for the challenge of identifying and addressing abusive language posted in these digital environments. Recent research has increasingly focused on evaluating the ability of machine-learning language algorithms to address hate speech. This entails examining the wider social implications of moderation practices on social media content and exploring community-centered research efforts to reduce bias in automated algorithms that disproportionately affect marginalized groups. 

Despite extensive research on English-language posts across social media platforms, hate-speech detection systems still struggle with implicit, non-pejorative, and reclaimed English language. This complicates accurate identification of abusive content aimed at individuals based on identity characteristics such as gender, race, ethnicity, or disability, especially when multiple aspects of a person’s identity are targeted simultaneously (e.g., women of color and LGBTQ+ community members with disabilities). To mitigate negative impacts of current automated content analysis and moderation practices, developers should incorporate more diverse, representative perspectives into training data and research methods used for developing hate-speech detection algorithms. 

Within the context of US-centered social media analysis, effective hate-speech detection algorithms are particularly important because of the rise in potentially abusive content associated with US national elections. A recent CDT report found that women of color political candidates faced a disproportionate amount of hate speech online in the summer leading up to the 2024 elections. That study used hate-speech detection algorithms to count the number of posts directed at candidates that contained hate speech. If anything, the report likely understated the problem since the flaws in these algorithms mean they under-identify hate speech. An earlier CDT report likewise found that in instances where political candidates are targeted, women of color experience more violent online harassment, reflecting the intersection of racist and misogynist sentiment. Hateful or abusive posts, especially when compounded by misinformation, contribute to a chilling effect that prompts many targeted individuals to self-censor and reduce their online presence.

Challenges in demystifying platform policies on hate speech partially stem from a lack of transparency around content moderation practices, including when and how they use AI systems. Major platforms like Instagram, Facebook, and X (formerly Twitter) use machine-learning algorithms to identify problematic content. Natural Language Processing (NLP), a subset of machine learning, is crucial for analyzing and generating text, with Large Language Models (LLMs) being particularly effective for reviewing social media content due to their ability to process large datasets and adapt to new instructions. This is a powerful feature for content moderators who are monitoring highly polarized and fast-paced digital environments on a daily basis. 

AI tools have transformed online content analysis and moderation, but biases in model training can lead systems to incorrectly flag posts, misidentify hate speech targets, and provide misleading context explanations (known as “hallucinations”). Choices made by researchers or moderators regarding data, annotation categories, and the quantity and types of examples used to prompt algorithms affect model performance and introduce algorithmic biases. Research shows that hate-speech detection often misidentifies or stereotypes individuals using dialects or group-specific language, like African American English (AAE) or LGBTQ+ vernacular, especially in contexts of language reclamation, self-identification, or educational content. 

Research also shows that automated content moderation on social media platforms often disadvantages marginalized groups. Despite Meta’s uniform hate speech policies, employees report that in practice Instagram’s automated algorithms detect more hate speech targeting white people than it detects targeting black people. Additionally, Black users also face a 50% higher likelihood of automatic account disabling. A study on intersectional hate speech found that only 17% of posts containing “misogynoir” (anti-Black forms of misogyny) on X were classified properly by popular hate-speech detection algorithms. Another study concluded that transgender Facebook users also face higher levels of removal of content when posting about their identities, such as by “coming out” or using terms like “queer” which is commonly reclaimed within the LGBTQ+ community. Studies on hate-speech detection language model performances have found that posts containing reclaimed terms are more likely to be considered harmful when used in counter speech, educational explanations, or discussions of self-identity by LGBTQ+ users. Independent research remains vital for understanding biases in automated hate-speech detection models. Yet support for external social media monitoring tools is limited, due to new data access restrictions from major platforms in the US.  

Expanding transparent, accessible, and community-driven research is essential for improving AI tools like automated content moderation and reducing bias in hate-speech detection. For example, group-specific approaches to LLM research center marginalized communities, and seek to understand how models respond to community-specific language use. Transparent and accessible research is vital for improving representation, and for building confidence in the merits of these approaches. Such research has shown, for example, that in automated hate-speech detection, LLM algorithms can quickly adapt to new language and concepts through additional prompting. 

Improving access to datasets is also a crucial step. A study reviewing hate-speech detection research found that, across published articles, 51% of the datasets used are publicly available, and less than 35% of datasets mention methodology. Tools cataloging hate speech datasets for public availability (e.g., Detoxify, QueerReclaimLex, HateSonar, and HateXplain to name a few) help to crowdsource research, offering opportunities to center expertise and collaboration across disciplines. I would also argue that integrating perspectives from content moderators is vital. Online community members can contribute to the review of content in accordance with platform rules (called “reactive moderation”).

Expanding diverse, community-driven research, improving method transparency, and increasing dataset access are crucial for developing fairer AI tools and mitigating adverse effects of current social media content moderation practices. Context analysis, fueled by interdisciplinary collaboration, enhances data collection quality, leading to more representative training of hate-speech detection models and stronger ethical considerations for people-centered research.