Inaugural Editorial

It is a particular kind of privilege to write the inaugural editorial of a new journal — notably because the writing is also the journal's first act of editorial position-taking. A first issue is both a beginning and an argument: the declaration that a field of inquiry has reached sufficient maturity to warrant a dedicated home, and that the community working in it deserves a venue commensurate with the urgency of its scholarship.

The argument the journal advances is twofold. The field of hate studies in digital space has developed conceptual, methodological, and empirical maturity sufficient to justify its own venue. At the same time, that maturity has not yet been matched by the field's internal cohesion, its interdisciplinary ambition, or its capacity to translate research into the policy, legal, and civic responses the problem demands. The first claim justifies the journal's launch; the second explains its design.

Research on online hate has proliferated across disciplines — computational linguistics, political communication, media studies, sociology, psychology, legal theory, Holocaust studies — producing work of genuine quality and social significance. Yet this proliferation has remained structurally fragmented. Computational studies appear in machine learning venues; discourse-analytic work in linguistics journals; sociological findings in political science outlets; historical scholarship in humanities publications. Scholars analyzing the linguistic mechanics of hate speech and the engineers building systems to detect it work on the same problem and rarely read each other, because of the structural gaps between disciplinary venues. Digital Hate Review is founded on the conviction that online hate studies is not merely a topic multiple disciplines happen to share, but an issue domain whose most important questions fall precisely into the blind spots no single discipline can see alone.

A fuller account of the journal's scope, scholarly rationale, and editorial standards is available on the Digital Hate Review website. What follows here introduces the journal's three-pillar architecture and the contributions that constitute Volume 1, Issue 1 across the Research, Legal Forum, and Perspectives sections.

Three Pillars, One Project

Digital Hate Review is structured around three pillars — distinct in form, unified in purpose.

Research Articles is the foundational pillar: a peer-reviewed research journal in the full and rigorous sense, publishing original empirical, theoretical, and methodological work on digital hate across all relevant disciplines, subject to double-blind peer review managed through ScholarOne. The Research section treats digital hate as an ecosystem phenomenon and methodological plurality as a disciplinary virtue, actively prioritizing work that brings together computational, qualitative, behavioral, legal, and large language model (LLM)-enhanced approaches rather than treating any single tradition as self-sufficient.

The Legal Forum is a structured, cumulative, citable scholarly space for legal and doctrinal engagement with digital hate, platform governance, and AI accountability. Its purpose is not to resolve doctrine or promote policy positions, but to clarify legal tensions, thresholds, and institutional responsibilities at the intersection of law, platforms, and artificial intelligence — to do so before those tensions harden into bad law. Legal experts are invited into a sustained, asynchronous, fully citable exchange on the journal's website — a format without a clear equivalent elsewhere — that allows positions to accumulate over time as a documented intellectual record. Conference panels are ephemeral; traditional journal articles are monological; the Legal Forum is neither.

Perspectives bridges scholarship and practice through editorially curated essays and conversations with researchers, platform engineers, civil society actors, policymakers, journalists, and public intellectuals working on platform dynamics and content governance. Perspectives provides a formal venue for practitioner and public-intellectual knowledge, which is unlike academic knowledge but no less important. By integrating these voices structurally into DHR's architecture rather than treating them as supplementary, the journal ensures the research it publishes is in a sincere conversation with the practical and operational realities it seeks to inform.

The Inaugural Issue: Research Articles

The assembled research articles reflect the journal's interdisciplinary mission. They address digital hate from computational, discourse-analytic, media-analytic, and journalism-standards perspectives, spanning multiple empirical contexts and forms of targeted hatred.

Mykola Makhortykh and Elizaveta Kuznetsova introduce and empirically ground the concept of "routine distortion": Holocaust misrepresentation that arises as an unprompted byproduct of generative AI's architecture, not from deliberate jailbreaking. Their intervention is to shift attention from high-profile viral cases to the lower-visibility but structurally more significant distortion that emerges in ordinary information-seeking. Survey data from Switzerland, the US, and Germany document that AI use for Holocaust-related questions is already substantial and growing rapidly. Empirical audits in parallel show AI applications answering correctly in only half to seventy percent of prompts at best — generating fictional war crime trials, invented testimonial quotes, and misattributed atrocities. The authors argue that because such distortions occur at the scale of the entire user base and appear highly convincing, the cumulative risk warrants attention comparable to that paid to viral cases.

Daniel Miehling analyzes affective polarization in over 3.4 million YouTube Shorts comments related to the 2023–2024 Israel-Hamas war, drawn from four state-funded outlets. Using a fine-tuned aspect-based sentiment model, the paper traces sentiment toward seven entities — Israel, Zionists, Palestine, Palestinians, Hamas, Jews, Muslims — finding negative sentiment toward Israel and Zionists stable and persistent across all outlets and the full twelve-month period. A complementary seedlist analysis documents the prevalence of Holocaust inversion framings and shows that "Zionist" functions primarily as a pejorative proxy, enabling hostility through ostensibly political vocabulary.

Patrick Y. Wu, Venkata Dhanush Kikkisetti, Suneela Maddineni, and Nathalie Japkowicz present the Hate Mitigation App (HMApp), a computational pipeline for the automated discovery of emergent coded and non-coded hate speech terminology on social media, tracking hate lexica across five categories — antisemitism, anti-Black hatred, Islamophobia, anti-Asian hatred, and anti-LGBTQ+ hatred. Their dual evaluation framework surfaces a sobering asymmetry: established lexicons capture only a fraction of the hate terminology actually circulating online. The finding signals a measurable gap between deployed moderation lexicons and the terminology now in circulation.

Karina Halevy, Julia Mendelsohn, Chan Young Park, Yulia Tsvetkov, and Maarten Sap introduce a task conceptually distinct from hate speech detection: hateful event detection — fine-grained classification of reports describing antisemitic incidents from news articles, civil society monitoring, and official records. The distinction matters: existing NLP approaches center on speech rather than events, systematically excluding bias-motivated physical violence, vandalism, and discrimination. Evaluating LLMs against a fine-grained taxonomy grounded in IHRA and Jerusalem Declaration definitions, the paper proposes fine-grained taxonomies as a route to both computational tractability and practitioner utility.

Tatiana Glezer analyzes 1,378 articles published between February and June 2024 by major English-language outlets and international news agencies, applying a systematic coding scheme to all casualty-related statements. The findings document measurable asymmetries: Hamas-run Gaza Ministry of Health figures cited in 100% of publications, Israeli data on militant casualties in 3%, and the essential methodological disclaimer that Gaza MoH figures do not distinguish civilians from combatants appearing in only 15% of articles overall. The paper argues that these asymmetries produce what it terms an "authoritative version of reality," and calls for explicit professional standards in casualty-data reporting.

Taken together, these five contributions enact rather than merely describe what Digital Hate Review endeavors to accomplish. Each engages a distinct disciplinary tradition — computational linguistics, NLP, AI auditing, computational social science, journalism studies — yet each depends, explicitly or implicitly, on what the others know. None could have appeared, without remainder, in an academic venue designed for a single discipline.

The Inaugural Issue: Legal Forum

The Legal Forum opens with one of the most acute doctrinal challenges in contemporary digital governance: the legal and regulatory frameworks for AI-generated and AI-amplified hate content, engaged across First Amendment doctrine and the Digital Services Act tradition. Sophie Xiaoyi Liu opens the thread, arguing that autonomous AI agents and decentralized multi-agent systems are producing a new category of digital harm — harassment, hate speech, and disinformation generated without any identifiable or legally accountable human decision-maker — that exposes a structural collapse in existing legal frameworks built around the assumption of a traceable anchor of responsibility. Her position is engaged by scholars across constitutional, comparative, and international legal traditions, and the thread closes with an editorial reading of the doctrinal questions the field cannot yet answer.

Topics the Legal Forum will engage in subsequent threads include transparency, accountability, and due process for platforms and AI systems; freedom of expression and the regulation of harm; responsible and accountable AI — from ethical commitments to legal responsibility; the balance between individual rights and the prevention of radicalization and violence; government–platform–AI company relations; platform governance and private power; the role of evidence, expertise, and interdisciplinary scholarship in legal decision-making; victims' rights and access to justice in digital and AI-mediated contexts; and comparative, global, and future-facing questions across jurisdictions. The Forum welcomes contributions from scholars in law, legal theory, digital governance, and adjacent fields.

The Inaugural Issue: Perspectives

Perspectives opens with a set of essays addressing precisely the points of friction, translation gaps, and implementation failures that are absent from purely academic discourse.

Max Shulman-Litwin (Cyabra) addresses the structural shift from message-based propaganda to identity-based manipulation. Generative AI has enabled the rise of "faux-fluencers" — highly realistic synthetic personas that build trust and parasocial attachment to spread scams, propaganda, and hate at scale. These AI-driven identities manipulate audiences through perceived authenticity, embedding persuasion within seemingly organic social interaction. The essay traces synthetic personas across financial fraud, medical misinformation, racist, antisemitic, and anti-LGBTQ+ content, and state-sponsored information warfare, and argues that current platform policies and legal frameworks are inadequate to the threat.

David Kuszmar (Gazzetta), in his respective essay, discusses what the platform/AI-mediated behaviors mean for the spread of digital hate: LLMs and AI chatbots function as deeply personalized systems of behavioral influence, mirroring users' worldviews to maximize engagement and producing what he terms a "mirrored cage" of hyper-tailored information environments in which beliefs and behaviors are subtly shaped. Kuszmar's argument points to a conclusion the field is still learning to absorb: countering digital hate may require approaches closer to breaking addictive behaviors than to fact-checking or policy intervention.

Mayank Kejriwal (University of Southern California) addresses the other side of the AI question — not how AI shapes user behavior, but how the resulting systems should be governed. Kejriwal addresses the accountability problem posed by agentic AI, arguing that systems capable of autonomous, multi-step actions across digital platforms must be inherently governable — monitorable, attributable, constrainable, and interruptible during operation. His proposed framework of "governability-by-design" rests on persistent identity tracking, auditable logs, bounded autonomy, human oversight, and cross-platform interoperability.

Across these essays, a range emerges that no single-discipline journal section could host: decentralized peer-to-peer extremism, the behavioral psychology of platform and AI use, and the governance architecture of agentic systems. Additional Perspectives essays and curated conversations will appear across subsequent issues.

A Note on Method, and on the Years Ahead

The peer-review principle is non-negotiable. But a journal that publishes excellent scholarship and leaves it stranded within disciplinary boundaries does not live up to the stakes of the problems it explores. The findings of digital hate studies bear directly on the safety of real-world communities, on the resilience of democratic institutions, and on the strength of the legal and educational architectures protecting them. Those stakes impose obligations on the journal: to be rigorous, to be integrative, to be honest about evidence, and to speak clearly to the audiences who need to hear what we know.

What is at stake is the quality of the public conversation democratic societies are capable of sustaining about the hardest problems of digital life. Digital Hate Review is launched in the belief that better scholarship, brought into honest contact with the actors who shape that conversation, can still change what the conversation contains.

The problem that motivates this journal — the proliferation and intensification of hate in digital spaces, and the structural obstacles that prevent knowledge about that problem from reaching those who most need it — will not resolve itself. Research provokes answers; the public sphere produces solutions; the work of a journal is to make sure the first reaches the second in time to matter.

What success looks like, in five years, is a journal whose work is cited not only by researchers but also by legislators drafting platform regulation, judges reasoning about AI liability, educators designing digital literacy curricula, and civil society organizations building counter-hate interventions. The inaugural issue is a beginning.

I am grateful to Digital Hate Review's Managing Editor, Keyu Glanz, for helping shape this first issue from its earliest stages. The rigor, judgment, and steady editorial work behind every page reflect his contributions as much as my own. I am also grateful to Emily Carmeli, President of AddressHate, and to the AddressHate core team for the institutional support that has made the launch of this journal possible—and made it feel, throughout, like a collective undertaking rather than a solitary one. Finally, I thank the Digital Hate Review Editorial Board for the scholarly guidance that will shape the journal's future, as well as the authors and reviewers whose work fills these pages.

Matthias J. Becker, PhD, Editor-in-Chief, Digital Hate Review