The hottest Content Moderation Substack posts right now

And their main takeaways
Category
Top News Topics

Gemini: how did we end up here?

lcamtuf’s thing • 2652 implied HN points • 02 Mar 24
🕹 Technology Content Moderation
  1. The development of large language models (LLMs) like Gemini involves mechanisms like reinforcement learning from human feedback, which can lead to biases and quirky responses.
  2. Concerns arise about the use of LLMs for automated content moderation and the potential impact on historical and political education for children.
  3. The shift within Big Tech towards paternalistic content moderation reflects a move away from the libertarian culture predominant until the mid-2010s, highlighting evolving perspectives on regulating information online.

BIG OOF: Patreon is enabling an insane stalking campaign

Natalia Mitigates The Apocalypse • 353 implied HN points • 29 Jan 24
🕹 Technology Content Moderation
  1. Stalking can happen to anyone, not just famous people, and tech companies like Patreon can inadvertently enable stalkers.
  2. Documenting and calling out instances of stalking and harassment can help raise awareness and hold tech companies accountable for their role.
  3. Coping strategies against online harassment include setting boundaries, exercising, practicing meditation, and using creativity to tell your story.

My Fake AI Problem

John’s Substack • 17 implied HN points • 05 Feb 26
🕹 Technology Content Moderation
  1. AI-generated fake videos can be so convincing that even people who know the subject well may be fooled.
  2. This is a widespread problem affecting many public figures, and platform enforcement struggles mean removing fakes often feels like a whack-a-mole effort.
  3. There may not be a clear solution yet, so everyone should stay alert and verify videos before trusting or sharing them.

Substackers Against Nazis

Japan Economy Watch • 399 implied HN points • 16 Dec 23
🇺🇸 U.S. Politics Content Moderation
  1. The Substack platform is being criticized for allowing the platforming and monetization of Nazis and white nationalists, which has caused concern among subscribers.
  2. Many prominent Substack writers have left or threatened to leave due to Substack's inability to adequately address the issue of allowing white nationalism on the platform.
  3. Subscribers and publishers are calling on Substack to clarify their stance on platforming Nazis and to reconsider their position on allowing such content to be monetized.

Elon Musks Degradation Engine

GOOD INTERNET • 37 implied HN points • 06 Jan 26
🕹 Technology Content Moderation
  1. A mainstream platform added a nudify feature that let an AI undress and sexualize people’s photos at scale, producing thousands of nonconsensual sexual images — including of minors.
  2. Turning sexual imagination into an automated publishing tool industrializes the male gaze, creating a constant swarm-like pressure that degrades women’s dignity and harms identity formation, especially for teenage girls.
  3. Enabling and monetizing this tool shows a disregard for privacy and dignity, and has provoked regulatory backlash, legal risks, and calls for bans or stronger enforcement.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

The existential problems with Threads

Hot Takes • 471 implied HN points • 07 Jul 23
🕹 Technology Content Moderation
  1. Threads faces challenges in attracting users away from established platforms due to oversaturation and user fatigue.
  2. The lack of financial incentives for users on Threads puts it at a disadvantage in a landscape where users value their time and content.
  3. Privacy concerns, trust issues, and the risk of censorship could deter users from joining Threads, impacting its success.

Bluesky follower sales websites

Conspirador Norteño • 52 implied HN points • 07 Dec 25
🕹 Technology Content Moderation
  1. Websites selling Bluesky followers, likes, and reposts have multiplied and are easy to find with a simple search as the platform grows.
  2. Many of those sites look nearly identical, use the same chat widgets (often backed by LLMs), and rely on similar hosting, which suggests shared operators or common tooling.
  3. Fake follower accounts show a repetitive bio pattern like “X based, interested in Y,” and thousands were created in bulk, indicating they were manufactured for sale.

Astroturfing where the skies are blue

Conspirador Norteño • 36 implied HN points • 20 Dec 25
🔮 Crypto Content Moderation
  1. A coordinated network of thousands of spam accounts massively amplified an "enterprise-grade Web3" post on Bluesky by reposting, liking, replying, and quote-posting to create fake popularity.
  2. Most of the accounts were created on the same two days, interacted mainly with the promoted account and a few crypto/music posts, and posted at very high rates — behavior consistent with a spam-for-hire operation that led to many suspensions.
  3. The promoted site offers AI-agent services that require payment in an obscure token ($MUSA), suggesting the astroturfing was intended to drive traffic and paid conversions to a crypto-linked service.

Twitter and Content Moderation

Life Since the Baby Boom • 461 implied HN points • 18 Feb 25
🇺🇸 U.S. Politics Content Moderation
  1. Elon Musk and Mark Zuckerberg have changed how fact-checking is done on their platforms. Instead of having official fact-checkers, they are relying on community input.
  2. Many suggested corrections on these platforms seem to just be people having different opinions. Often, they don't provide clear facts and can be more about arguing than informing.
  3. The interaction with these community notes doesn't seem to attract much attention. Most users appear to prefer the regular replies over community notes, which might indicate a lack of interest in these corrections.

Facebook's Challenges

In My Tribe • 379 implied HN points • 25 Oct 24
🕹 Technology Content Moderation
  1. Facebook struggles with content moderation because it has to balance user complaints. If they are too strict or too lenient, someone will be unhappy.
  2. Switching to a subscription model would likely not work well for Facebook since it would lose valuable user data that helps target ads.
  3. Facebook sees TikTok as a competitor and has changed its platform to reach users who want to connect with strangers, which has led to some issues with political content.

Here's how to stop the harmful content before it's posted

techandsocialcohesion • 59 implied HN points • 29 Mar 24
🕹 Technology Content Moderation
  1. Researchers are exploring using AI to prevent toxic content before it's posted online by prompting users as they type messages.
  2. Users appreciated the concept of receiving alerts about potentially harmful language but had concerns about privacy and disruptions to natural conversation flow.
  3. Implementing proactive measures like AI-based content moderation prompts not only eases the burden on moderation systems but also enhances the quality of online interactions by promoting empathy and understanding.

One AI-generated human race, part II

Conspirador Norteño • 16 implied HN points • 14 Dec 25
🕹 Technology Content Moderation
  1. A coordinated TikTok spam network of at least 76 accounts posts highly repetitive AI-generated videos of nonexistent people, and many clips show obvious AI glitches.
  2. The network’s content has broadened from friendly messages to recurring themes like romantic couple scenes, dating prompts, and staged emergency workers, all using similar fake people and backgrounds.
  3. Many accounts later pivot to commercial spam—mainly dietary supplements in English and Spanish with Paid Partnership labels—suggesting the AI videos were used to farm engagement before trying to monetize.

Mark Zuckerberg's commitment to free speech is as deep as Exxon's commitment to clean energy

The Future, Now and Then • 198 implied HN points • 09 Jan 25
🇺🇸 U.S. Politics Content Moderation
  1. Meta's commitment to free speech and content moderation is often temporary and depends on political convenience. They respond to crises when pressured but quickly revert to leniency when the heat is off.
  2. Zuckerberg's recent shift in moderation policies suggests a move towards cheaper, less effective community-driven solutions instead of rigorous fact-checking, potentially increasing harm towards marginalized groups.
  3. There's a growing debate about whether big tech companies deserve special treatment from the government, with some arguing that they should follow the same rules as everyone else despite their influence.

The Work of Free Speech on Social Media

Oliver Bateman Does the Work • 98 implied HN points • 14 Dec 23
🇺🇸 U.S. Politics Content Moderation
  1. The banning of certain figures on social media is often driven by public relations considerations rather than purely moral or ethical standards.
  2. Social media platforms prioritize maintaining a certain inoffensive public image to attract ad revenue and align with mainstream media-approved trends.
  3. The dynamics of speech regulation on social media platforms spark debates about freedom of speech, corporate interests, and user empowerment in the digital age.

✅ The great social media migration

Wadds Inc. newsletter • 2 HN points • 02 Sep 24
🕹 Technology Content Moderation
  1. Many users are frustrated with X due to misinformation and toxic conversations, pushing them to consider other platforms. It's a tough decision since X has been important for news and networking.
  2. Some companies and professionals are staying quiet on X, making it hard to leave a platform that has been so integral to their work and connections.
  3. There's a growing interest in new platforms like Threads, which are trying to offer more decentralized and user-controlled social media experiences.

The curious case of "TikTok refugees" on Rednote/Xiaohongshu

Pekingnology • 158 implied HN points • 14 Jan 25
🕹 Technology Content Moderation
  1. Many TikTok users in the U.S. are moving to a Chinese app called RedNote due to fears of a TikTok ban. This has led to an increase in the app's popularity.
  2. RedNote is like a mix of TikTok and Instagram, mainly used by young people to share lifestyle tips. However, it hasn't been widely known outside of Chinese-speaking areas until now.
  3. The move raises concerns about content moderation and privacy. RedNote may struggle with foreign-language content and could face pressure from Chinese regulations as more American users join.

Bluesky, smiling at me

WORLD GONE WRONG • 98 implied HN points • 13 May 23
🕹 Technology Content Moderation
  1. Bluesky is a small social app with positive vibes but lacks universal moderation.
  2. Bluesky's content filtering allows users to manage what they see based on categories.
  3. Content moderation is effective at limiting how bad actors exploit online platforms.

Ghosting Substack

Alex's Personal Blog • 361 implied HN points • 12 Jan 24
🕹 Technology Content Moderation
  1. Substack's content moderation decisions impact its users and the platform's image.
  2. Allowing Nazis on a platform can drive away non-Nazi users.
  3. Seeking alternative platforms that align with personal values is a valid response to disagreements with moderation policies.

Bullet Points: OpenAI looks for its FarmVille, 90s tech nostalgia, and more on the Substack drama

The Future, Now and Then • 275 implied HN points • 13 Jan 24
🕹 Technology Content Moderation
  1. OpenAI is aiming to become a platform similar to how Facebook invited developers for apps.
  2. Nostalgia for 90s tech optimism is prevalent but may not be constructive for the present tech landscape.
  3. The Substack management issues are conflicting with their core value of empowering writers to build their own audience.

Artificial Division -

Charles Eisenstein • 2 implied HN points • 30 Jan 26
🎭️ Culture Content Moderation
  1. Comments will be opened but limited to paid subscribers to set a boundary, and anyone who sincerely asks can get a free upgrade so money isn’t a barrier.
  2. The aim is to allow critique of ideas while blocking personal attacks, derision, and cruelty to keep discussion constructive and safe.
  3. There’s a community forum on Mighty Networks with a live call coming up, and the Substack is reader-supported with options to subscribe.

What's the difference between a Nazi and a pornographer?

reedmolbak • 19 implied HN points • 30 Dec 23
🇺🇸 U.S. Politics Content Moderation
  1. Content moderation is a complex issue for platforms like Substack, with controversies around what kind of content should or should not be allowed.
  2. Substack's moderation policies are driven by a mix of factors, including branding, promotion of free speech, and distinguishing between harmful content and open discourse of ideas.
  3. There's a fine line between allowing free speech and banning harmful content, and platforms like Substack make value judgments based on what is considered harmful or not communicating ideas.

Zoom event Monday evening October 14

In My Tribe • 45 implied HN points • 11 Oct 24
🇺🇸 U.S. Politics Content Moderation
  1. There's a Zoom event on October 14 at 8 PM New York time with John Samples discussing content moderation on Facebook.
  2. The event will also touch on the current state of political conflict and where it might be headed.
  3. This event is exclusive to paid subscribers, so make sure to sign up if you want to join.

AI Gone Rogue... in a Good Way!

aidaily • 19 implied HN points • 17 Aug 23
🕹 Technology Content Moderation
  1. The military has a team called Generative AI Task Force to make AI creative.
  2. Netflix and Walmart are offering high salaries to AI experts, sparking a job feud.
  3. People are divided on whether AI should be involved in voting, questioning its reliability.

What did Elon do?

Random Minds by Katherine Brodsky • 126 implied HN points • 10 Apr 23
🕹 Technology Content Moderation
  1. Users of Substack had issues posting links and engaging with content, mirroring past actions by Elon Musk.
  2. Search bans were placed on accounts and terms, leading to difficulties in accessing Twitter content.
  3. The situation raised questions about transparency, trust in Elon Musk, and the fairness of link restrictions on Twitter.

Áo giáp và thanh gươm

Thái | Hacker | Kỹ sư tin tặc • 59 implied HN points • 20 Jul 20
🕹 Technology Content Moderation
  1. Growing up with the internet has shaped the author's worldview, encouraging critical thinking and curiosity that traditional schooling may not have fostered.
  2. Access to the internet has provided the author with valuable knowledge, career opportunities beyond Vietnam's borders, and a broader understanding of the impact of the internet on society.
  3. The Section 230 law in the US, granting websites immunity for content posted by users, has played a significant role in the development of the internet industry, emphasizing the balance between freedom, responsibility, and innovation.

Instagram and Threads making Politics opt-in is fine

GOOD INTERNET • 10 implied HN points • 12 Feb 24
🕹 Technology Content Moderation
  1. Meta will no longer recommend political content across its apps like Instagram and Threads, aiming to create a more apolitical social media environment.
  2. It's essential to recognize the shift towards making divisive political content opt-in rather than default on social media platforms.
  3. While this move may limit exposure to important topics like climate action, it reflects an attempt to make social media platforms more like a pub, avoiding contentious political discussions.

Elon Musk is Exactly Why I Don't Want Twitter Heavily Curating Content

I Might Be Wrong • 10 implied HN points • 12 Apr 23
🇺🇸 U.S. Politics Content Moderation
  1. Progressives advocate for more content curation on social media to combat hate speech and misinformation.
  2. Elon Musk's leadership of Twitter following progressive demands showcases challenges with content moderation.
  3. Having large social media companies control speech raises concerns over power and influence in our national dialogue.