The Gray Area: Characterizing Moderator Disagreement on Reddit

TL;DR

This study introduces the concept of the moderation gray area on Reddit, defined as cases where multiple moderators dispute a single target decision. Using a 5-year, 24-subreddit ModLog dataset, the authors quantify that roughly 14% of moderation actions reside in the gray area, with bots disproportionately initiating removals later overturned by humans. They dissect the gray area through action sequences, moderator rationales, and content topics, revealing that bot-driven overmoderation and expert human interventions are central to final outcomes, while automated systems struggle with dispute content. Through information-theoretic analysis and LLM benchmarking, they show gray-area cases are inherently harder to adjudicate than undisputed ones, with current models underperforming on disputed content; they argue for stronger human-in-the-loop designs and participatory governance to improve moderation quality and fairness.

Abstract

Volunteer moderators play a crucial role in sustaining online dialogue, but they often disagree about what should or should not be allowed. In this paper, we study the complexity of content moderation with a focus on disagreements between moderators, which we term the ``gray area'' of moderation. Leveraging 5 years and 4.3 million moderation log entries from 24 subreddits of different topics and sizes, we characterize how gray area, or disputed cases, differ from undisputed cases. We show that one-in-seven moderation cases are disputed among moderators, often addressing transgressions where users' intent is not directly legible, such as in trolling and brigading, as well as tensions around community governance. This is concerning, as almost half of all gray area cases involved automated moderation decisions. Through information-theoretic evaluations, we demonstrate that gray area cases are inherently harder to adjudicate than undisputed cases and show that state-of-the-art language models struggle to adjudicate them. We highlight the key role of expert human moderators in overseeing the moderation process and provide insights about the challenges of current moderation processes and tools.

Figures (11)

• Figure 1: (a) An example showing multiple moderation actions on a single comment. (b) Cases are partitioned into four mutually exclusive strata by (i) whether there is within-case disagreement defined as more than one unique moderator and more than one unique action and (ii) whether any moderator is a bot. The case (a) is categorized into "gray bot".
• Figure 2: Share of moderation cases by stratum (gray vs. undisputed) across rule categories. Gray-area cases are relatively overrepresented in trolling, brigading, and doxxing, while spam, link-only, and formatting violations make up a larger share of undisputed cases.
• Figure 3: Differences in text difficulty (PVI) relative to the "undisputed human" baseline, with positive values indicating clearer (easier) text and negative values indicating more ambiguous (harder) text across quantiles. Shaded areas represent pointwise 95% bootstrap confidence intervals.
• Figure 4: Macro-F1 and 95% confidence intervals for different models across case strata.
• Figure 5: Distribution of moderators’ join dates in subreddits, used with action timestamps to calculate moderator experience (in days).