Tag
gg-friggin-ez is an open-source Node.js tool for fast, multilingual profanity and toxicity screening, supporting languages like Hindi and Bengali with evasion-aware features.
The author discusses the growing pains of a subreddit focused on large language models and proposes rules to promote science-based discussions and reduce politics and misinformation.
The author questions why verified social media accounts are posting spam despite a paid verification process, highlighting issues with online moderation.
The article discusses modifications to the Claude AI model that were disliked or not approved by moderators, expressed with uncertainty.
The Consumer Rights Wiki announces updates to its platform, including new editor feedback features, anti-spam measures, MediaWiki upgrades, and enhanced moderator tools.
A Reddit user expresses frustration with bots spamming AI-generated content and suggests restricting new or low-karma accounts to improve the community.
A lobste.rs user argues that the 'vibecoding' tag is being over-applied to posts that are actually the product of careful human effort, and calls for a discussion about the tag's nuances.
Elon Musk criticizes Twitch for an alleged double standard in moderation, citing Hasan Piker's past comments versus Asmongold's ban.
Reddit signals upcoming changes to old.reddit.com and starts restricting new API requests, pushing third-party apps to its Developer Platform, continuing its contentious relationship with developers and AI companies.
Mistral releases Shieldstral-1.0-3B, a compact safety/moderator model jokingly referenced as 'Not-Hotdog', aimed at filtering harmful content.
The Rust project is adopting a new policy governing how LLMs can be used when contributing to the rust-lang/rust monorepo, formalizing rules for PR authors, reviewers, and issue reporters.
PieFed, a federated social media platform, highlights its anti-fascist features for community moderation.
Telegram CEO Pavel Durov claims an extortionist planted AI-modified CSAM in a public chat to get Telegram temporarily removed from Apple's App Store, warning that this creates a systemic risk for all user-generated content apps.
Reddit has become a prime target for AI-driven SEO spam as brands use fake accounts and bots to covertly promote products, capitalizing on AI chatbots' tendency to cite Reddit. Moderators are now cracking down on these promotional tactics.
The article discusses AI-generated images posted on social media that fool both users and moderators, highlighting how platforms fail to label or flag AI content and actively obfuscate its origin.
LinkedIn is adding a button to let users report AI-generated 'slop' on its platform, as part of broader efforts to curb low-quality AI content.
An article discussing allegations of racism on the character.ai platform, highlighting concerns about AI moderation and ethical AI deployment.
A data breach at SunoAI is being discussed on Discord, with moderators issuing timeouts to users who bring up the topic.
Goose, a gay dating app marketed as friendship-focused, faces criticism over moderation and feelings of exclusion among users of color, while still being used for hookups.
An analysis of arXiv submission delay times since 2015 finds no recent increase in overall delays, despite moderation policy changes and increased submissions.