AI uBlock Blacklist
This GitHub project offers an uBlock Origin blacklist specifically targeting AI-generated content farms, providing a filter against the growing wave of low-quality, AI-produced web content. It resonates deeply with Hacker News users frustrated by SEO-optimized digital noise that lacks genuine human insight and clogs search results. The project's blunt 'Cry about it' FAQ response, while sparking debate, encapsulates a defiant stance many appreciate in the ongoing battle for a higher-quality internet.
The Lowdown
The GitHub project "AI uBlock Blacklist" by alvi-se offers a custom filter list for uBlock Origin, designed to block websites primarily generated by AI. The author's motivation stems from a personal frustration with encountering "mediocre content" filled with ads and referral links that provide "no useful information" and can be potentially dangerous due to factual inaccuracies or "hallucinations." The core premise is that users seeking information online desire answers written by a human with genuine experience and creativity, not an AI.
Key aspects of the project include:
- Installation: Users can subscribe to the list via a direct link for uBlock Origin or manually import the provided URL.
- Rationale: AI content farms are viewed as useless, ad-laden, potentially harmful (e.g., suggesting dangerous actions), and devoid of human value. The list aims to restore human-authored content to search results.
- Manual Curation: Websites are added manually by the author based on personal browsing encounters. While pull requests are welcomed, the author notes a bias towards Italian websites due to their nationality. Automated tools for detection are avoided due to their difficulty in accurately assessing content quality.
- Detection Patterns: The project outlines numerous characteristics used to identify AI-generated content, such as:
- Overly baroque or unnecessary introductions and conclusions.
- Generic titles like "A Comprehensive Guide" or "Ultimate Guide."
- Lack of external links, sources, or references.
- Ubiquitous referral links and product promotions.
- Blogs publishing hundreds of articles rapidly, often under the same author.
- Hallucinations or factually incorrect information.
- Publication dates after November 2022 (post-ChatGPT release).
- Absence of diverse media (images, videos) or presence of AI-generated media.
- Poor text formatting or un-rendered Markdown.
- Long posts with irrelevant content.
- Aggressive SEO tactics leading to high search engine rankings.
- "Know-it-all" blogs covering disparate topics.
- Vague content, and missing or unprofessional contact information.
- "AI enthusiastic" content on dedicated AI blogs.
- Google Dorks: The project suggests using specific phrases (e.g., "Sure! Here's an article about") that LLMs often use in their introductions to find 100% AI-generated pages for inclusion.
- "Never Asked Questions" (NAQ): This section features a notably defiant stance on websites included in the list, including "My website is on your list! Cry about it." and dismissing grammar correction via AI as a "skill issue."
While the blacklist offers a direct solution to a perceived degradation of web content, its maintainer's uncompromising attitude, particularly regarding delisting, has sparked debate regarding the long-term viability and fairness of such manually curated blacklists in the broader online ecosystem.
The Gossip
Blacklist Brashness & Bias
The maintainer's unapologetic "Cry about it" stance in the project's FAQ ignited significant discussion. Critics argued that such an attitude is problematic for a public blacklist, citing concerns about potential false positives, the absence of a removal process, and the reputational damage it could inflict on domain owners, drawing parallels to other content blocklists. Conversely, some commenters, while noting the creator's bluntness, expressed support for the frustration driving the list's creation.
The 'Skill Issue' & AI Writing Aids
The project's dismissal of using AI for grammar correction as a "skill issue" prompted a tangential discussion on the legitimate applications of AI writing tools. Users debated how tools like Grammarly assist non-native English speakers or individuals with communication challenges, contrasting this beneficial use with the "AI slop" the blacklist targets. Many acknowledged that admitting a "skill issue" is precisely why such tools are adopted.
Alternative Blocking & Detection Strategies
Commenters shared various other tools, lists, and conceptual approaches for identifying and blocking AI-generated or low-quality web content. Suggestions included existing uBlock Origin lists, specialized AI detection tools, and ideas for creating 'green lists' (whitelists). Some noted the distinct focus of this project on 'content farms' compared to broader anti-AI lists.
AI's Evolutionary Arms Race
Several users discussed the