#content-moderation — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #content-moderation, aggregated by home.social.
-
Censorship Lawsuit Against Disinformation Researchers Dismissed — After Succeeding At Actual Censorship
-
Censorship Lawsuit Against Disinformation Researchers Dismissed — After Succeeding At Actual Censorship
-
The Conversation: If we force online platforms to control harmful content, where does that leave sex ed?. “Current online safety rules are focused on removing harmful content, not on supporting health promotion. Unfortunately, sexual health content is often flagged as ‘against community standards’ and suppressed by platforms – a practice known as shadowbanning. But Australia’s promised new […]
https://rbfirehose.com/2026/07/04/the-conversation-if-we-force-online-platforms-to-control-harmful-content-where-does-that-leave-sex-ed/ -
The Conversation: If we force online platforms to control harmful content, where does that leave sex ed?. “Current online safety rules are focused on removing harmful content, not on supporting health promotion. Unfortunately, sexual health content is often flagged as ‘against community standards’ and suppressed by platforms – a practice known as shadowbanning. But Australia’s promised new […]
https://rbfirehose.com/2026/07/04/the-conversation-if-we-force-online-platforms-to-control-harmful-content-where-does-that-leave-sex-ed/ -
Ctrl-Alt-Speech: Making The Best Of A Ban Situation
-
Ctrl-Alt-Speech: Making The Best Of A Ban Situation
-
Moderation Without Memory
By Cliff Potts, CSO, and Editor-in-Chief of WPS News
Baybay City, Leyte, Philippines — July 2, 2026
Enforcement That Forgets Itself
One of the most damaging features of TikTok’s governance is not simply inconsistency. It is amnesia.
Moderation on TikTok does not appear to accumulate institutional memory. Each enforcement action behaves as if it exists in isolation, disconnected from prior decisions, patterns, or context. Users are not judged by stable standards. They are judged by moments.
This matters, because systems without memory cannot be fair. They can only be reactive.
Rules Without Precedent
In mature regulatory systems, precedent matters. Past decisions inform future ones. Patterns are recognized. Errors are corrected. Standards evolve visibly.
On TikTok, enforcement offers none of that continuity.
Creators routinely encounter:
- Content removed for reasons that contradict earlier approvals
- Penalties applied to one account but ignored on another for identical material
- No explanation of how prior compliance is weighed
- No visible record of how decisions are made over time
Without precedent, users cannot learn. Without learning, compliance becomes guesswork.
This is not merely frustrating. It is destabilizing (Gillespie, 2018).
Automation Without Accountability
Much of TikTok’s moderation appears to be automated or semi-automated, with minimal human review. Automation at scale is unavoidable. Unaccountable automation is not.
When automated systems operate without transparency or institutional memory, they amplify randomness. They do not enforce norms. They generate noise (Pasquale, 2015).
Appeals rarely correct this problem. They are often handled by the same systems that generated the original action, producing circular outcomes that feel less like review and more like confirmation.
A system cannot meaningfully review itself if it does not remember what it has done before.
The Human Cost of Forgetfulness
Moderation without memory changes how people behave.
Creators stop building archives. Sellers hesitate to invest in durable content. Users learn that nothing on the platform is permanent—not because of creative impermanence, but because governance itself is unstable.
This produces a short-term culture:
- Trends over substance
- Virality over reliability
- Disposable content over durable work
That culture may inflate engagement metrics. It erodes trust.
Why Memory Matters for Commerce
Commerce depends on continuity.
Sellers need to know that past compliance reduces future risk. Consumers need confidence that legitimate vendors will not vanish without explanation. Platforms facilitating transactions must be able to explain why one action occurred today and how it relates to yesterday.
A moderation system that forgets itself cannot do that.
This is not a minor flaw. It is a structural incompatibility between TikTok’s governance model and the requirements of serious commerce (Srnicek, 2017; Zuboff, 2019).
The Pattern, Not the Incident
This essay is not about a single takedown or a single mistake. It is about recognizing a recurring condition.
A platform that cannot remember its own decisions cannot be trusted to resolve disputes, protect participants, or enforce standards consistently. Over time, that failure becomes indistinguishable from indifference.
TikTok’s moderation system does not merely forget users. It forgets itself.
For more social commentary and excellent fiction, see Occupy 2.5 at https://Occupy25.com
This essay will be archived to the WPS News Monthly Brief available through Amazon.
APA Citations:
Gillespie, T. (2018). Custodians of the internet: Platforms, content moderation, and the hidden decisions that shape social media. Yale University Press.
Pasquale, F. (2015). The black box society: The secret algorithms that control money and information. Harvard University Press.
Srnicek, N. (2017). Platform capitalism. Polity Press.
Zuboff, S. (2019). The age of surveillance capitalism. PublicAffairs.
#algorithmicEnforcement #contentModeration #digitalCommerce #platformGovernance #socialMediaAccountability #TikTok -
Moderation Without Memory
By Cliff Potts, CSO, and Editor-in-Chief of WPS News
Baybay City, Leyte, Philippines — July 2, 2026
Enforcement That Forgets Itself
One of the most damaging features of TikTok’s governance is not simply inconsistency. It is amnesia.
Moderation on TikTok does not appear to accumulate institutional memory. Each enforcement action behaves as if it exists in isolation, disconnected from prior decisions, patterns, or context. Users are not judged by stable standards. They are judged by moments.
This matters, because systems without memory cannot be fair. They can only be reactive.
Rules Without Precedent
In mature regulatory systems, precedent matters. Past decisions inform future ones. Patterns are recognized. Errors are corrected. Standards evolve visibly.
On TikTok, enforcement offers none of that continuity.
Creators routinely encounter:
- Content removed for reasons that contradict earlier approvals
- Penalties applied to one account but ignored on another for identical material
- No explanation of how prior compliance is weighed
- No visible record of how decisions are made over time
Without precedent, users cannot learn. Without learning, compliance becomes guesswork.
This is not merely frustrating. It is destabilizing (Gillespie, 2018).
Automation Without Accountability
Much of TikTok’s moderation appears to be automated or semi-automated, with minimal human review. Automation at scale is unavoidable. Unaccountable automation is not.
When automated systems operate without transparency or institutional memory, they amplify randomness. They do not enforce norms. They generate noise (Pasquale, 2015).
Appeals rarely correct this problem. They are often handled by the same systems that generated the original action, producing circular outcomes that feel less like review and more like confirmation.
A system cannot meaningfully review itself if it does not remember what it has done before.
The Human Cost of Forgetfulness
Moderation without memory changes how people behave.
Creators stop building archives. Sellers hesitate to invest in durable content. Users learn that nothing on the platform is permanent—not because of creative impermanence, but because governance itself is unstable.
This produces a short-term culture:
- Trends over substance
- Virality over reliability
- Disposable content over durable work
That culture may inflate engagement metrics. It erodes trust.
Why Memory Matters for Commerce
Commerce depends on continuity.
Sellers need to know that past compliance reduces future risk. Consumers need confidence that legitimate vendors will not vanish without explanation. Platforms facilitating transactions must be able to explain why one action occurred today and how it relates to yesterday.
A moderation system that forgets itself cannot do that.
This is not a minor flaw. It is a structural incompatibility between TikTok’s governance model and the requirements of serious commerce (Srnicek, 2017; Zuboff, 2019).
The Pattern, Not the Incident
This essay is not about a single takedown or a single mistake. It is about recognizing a recurring condition.
A platform that cannot remember its own decisions cannot be trusted to resolve disputes, protect participants, or enforce standards consistently. Over time, that failure becomes indistinguishable from indifference.
TikTok’s moderation system does not merely forget users. It forgets itself.
For more social commentary and excellent fiction, see Occupy 2.5 at https://Occupy25.com
This essay will be archived to the WPS News Monthly Brief available through Amazon.
APA Citations:
Gillespie, T. (2018). Custodians of the internet: Platforms, content moderation, and the hidden decisions that shape social media. Yale University Press.
Pasquale, F. (2015). The black box society: The secret algorithms that control money and information. Harvard University Press.
Srnicek, N. (2017). Platform capitalism. Polity Press.
Zuboff, S. (2019). The age of surveillance capitalism. PublicAffairs.
#algorithmicEnforcement #contentModeration #digitalCommerce #platformGovernance #socialMediaAccountability #TikTok -
Gizmodo: One of Wikipedia’s Cofounders Banned From the Site Over Influence Campaigns. “After years of criticism and complaints about the direction of the site, Larry Sanger, one of Wikipedia’s cofounders, has been banned from editing articles. Sanger was indefinitely blocked from editing privileges on Wikipedia after volunteer editors accused him of violating the site’s rules by trying to […]
https://rbfirehose.com/2026/06/28/gizmodo-one-of-wikipedias-cofounders-banned-from-the-site-over-influence-campaigns/ -
Gizmodo: One of Wikipedia’s Cofounders Banned From the Site Over Influence Campaigns. “After years of criticism and complaints about the direction of the site, Larry Sanger, one of Wikipedia’s cofounders, has been banned from editing articles. Sanger was indefinitely blocked from editing privileges on Wikipedia after volunteer editors accused him of violating the site’s rules by trying to […]
https://rbfirehose.com/2026/06/28/gizmodo-one-of-wikipedias-cofounders-banned-from-the-site-over-influence-campaigns/ -
Ctrl-Alt-Speech: The Ctrl-Alt-Speech Reading List (Teaser)
-
Ctrl-Alt-Speech: The Ctrl-Alt-Speech Reading List (Teaser)
-
Judge Says Florida’s Social Media Law Is “Literally Impossible” To Obey. Thanks To The Supreme Court, It Gets A Trial Anyway.
-
Judge Says Florida’s Social Media Law Is “Literally Impossible” To Obey. Thanks To The Supreme Court, It Gets A Trial Anyway.
-
Ctrl-Alt-Speech: Close Your Apps And Think Of England
-
Ctrl-Alt-Speech: Close Your Apps And Think Of England
-
Instagram now has a preemptive censorship feature that warns users about making comments before they submit them. Because they want a "supportive place" i.e. you're not allowed to talk about anything controversial.
In this screenshot is a comment that "both the Dems and the Rs are corrupt" and that "if Dems are okay with genocide, what else are they okay with?"
Whether or not you agree with the statement, they are silencing trivial dissent, and that in-itself is revealing about our present and future.
I want folks to understand content moderation can be and is weaponized, and that content moderation is not a solved problem even on the #fediverse
#Instagram #meta #censorship #ContentModeration #Democrats #genocide #butchware
-
Instagram now has a preemptive censorship feature that warns users about making comments before they submit them. Because they want a "supportive place" i.e. you're not allowed to talk about anything controversial.
In this screenshot is a comment that "both the Dems and the Rs are corrupt" and that "if Dems are okay with genocide, what else are they okay with?"
Whether or not you agree with the statement, they are silencing trivial dissent, and that in-itself is revealing about our present and future.
I want folks to understand content moderation can be and is weaponized, and that content moderation is not a solved problem even on the #fediverse
#Instagram #meta #censorship #ContentModeration #Democrats #genocide #butchware
-
ChatGPT Spontaneously Generates Sexual Violence and Hardcore Snuff Imagery
https://mindgard.ai/blog/chatgpt-spontaneously-generated-violent-images-from-a-viral-prompt
#HackerNews #ChatGPT #SexualViolence #AI #Ethics #ContentModeration #MindGard
-
ChatGPT Spontaneously Generates Sexual Violence and Hardcore Snuff Imagery
https://mindgard.ai/blog/chatgpt-spontaneously-generated-violent-images-from-a-viral-prompt
#HackerNews #ChatGPT #SexualViolence #AI #Ethics #ContentModeration #MindGard
-
Ars Technica: Racist comments targeting politicians tripled since Meta relaxed its rules. “The researchers analyzed about 8 million Facebook comments and found that abusive and racist comments targeting both Republican and Democrat lawmakers tripled in the six months after the new rules were put in place. Some categories of abusive comments documented by the researchers saw even sharper rises, […]
https://rbfirehose.com/2026/06/13/ars-technica-racist-comments-targeting-politicians-tripled-since-meta-relaxed-its-rules/ -
Ars Technica: Racist comments targeting politicians tripled since Meta relaxed its rules. “The researchers analyzed about 8 million Facebook comments and found that abusive and racist comments targeting both Republican and Democrat lawmakers tripled in the six months after the new rules were put in place. Some categories of abusive comments documented by the researchers saw even sharper rises, […]
https://rbfirehose.com/2026/06/13/ars-technica-racist-comments-targeting-politicians-tripled-since-meta-relaxed-its-rules/ -
Trust & safety isn’t just cleaning up after harm — it’s protecting people before harm happens. Yoel Roth, SVP at Match Group, shares how AI tools and proactive tech are reshaping how platforms keep us safe. Curious about the future of digital safety? Dive in with us. #TrustAndSafety #DigitalSafety #ContentModeration
Watch on YouTube ▶️ https://youtube.com/shorts/9qNbBohgvpo
-
Anthropic Offers Mythos Upgrade for Cyber Partners and a ‘Safe’ Version for the Rest of You
http://web.brid.gy/r/https://www.wired.com/story/anthropic-releases-claude-fable-5-mythos-5/
-
Anthropic Offers Mythos Upgrade for Cyber Partners and a ‘Safe’ Version for the Rest of You
https://fed.brid.gy/r/https://www.wired.com/story/anthropic-releases-claude-fable-5-mythos-5/
-
TRUTH SOCIAL NAVIGATES MODERATION, ECHOES MEDIA CONCERNS
Truth Social claims to fight censorship but its moderation approach is being questioned by media experts. See what it means for users.
#TruthSocial, #ContentModeration, #SocialMedia, #MediaEthics, #CensorshipDebate
https://newsletter.tf/truth-social-moderation-policies-media-concerns/
-
TRUTH SOCIAL NAVIGATES MODERATION, ECHOES MEDIA CONCERNS
Truth Social claims to fight censorship but its moderation approach is being questioned by media experts. See what it means for users.
#TruthSocial, #ContentModeration, #SocialMedia, #MediaEthics, #CensorshipDebate
https://newsletter.tf/truth-social-moderation-policies-media-concerns/
-
Truth Social's content moderation policies are facing scrutiny, with some experts saying they are not as different from other platforms as claimed.
#TruthSocial, #ContentModeration, #SocialMedia, #MediaEthics, #CensorshipDebate
https://newsletter.tf/truth-social-moderation-policies-media-concerns/ -
Truth Social's content moderation policies are facing scrutiny, with some experts saying they are not as different from other platforms as claimed.
#TruthSocial, #ContentModeration, #SocialMedia, #MediaEthics, #CensorshipDebate
https://newsletter.tf/truth-social-moderation-policies-media-concerns/ -
ICYMI: AI-generated antisemitism hit 30 million views as platforms fall behind: CyberWell's report found 307 AI-generated antisemitic posts hit 30M views in 13 months. TikTok removed 88%, YouTube only 28%, and X just 20% of flagged content. https://ppc.land/ai-generated-antisemitism-hit-30-million-views-as-platforms-fall-behind/ #AI #Antisemitism #CyberWell #SocialMedia #ContentModeration
-
ICYMI: AI-generated antisemitism hit 30 million views as platforms fall behind: CyberWell's report found 307 AI-generated antisemitic posts hit 30M views in 13 months. TikTok removed 88%, YouTube only 28%, and X just 20% of flagged content. https://ppc.land/ai-generated-antisemitism-hit-30-million-views-as-platforms-fall-behind/ #AI #Antisemitism #CyberWell #SocialMedia #ContentModeration
-
Legislative Rhetoric Collides with Platform Governance
How do YouTube's May 2026 monetization rules affect political channels? Learn how community guidelines impact content creators and political discourse.
#youtubeupdate, #contentmoderation, #politicalnews, #digitalpolicy, #creatornews
-
YouTube is now reviewing entire channels instead of single videos to decide on monetization. This is a stricter approach than the rules used in 2025.
#youtubeupdate, #contentmoderation, #politicalnews, #digitalpolicy, #creatornews
https://newsletter.tf/youtube-monetization-rules-may-2026/ -
📢 CFP for our #MuC2026 workshop “Human-Centered #ContentModeration: Expertise, Context, and Evaluation” (30 Aug 2026)!
Submit:
📄 Position papers (2 pages)
✍️ Reflection essays (1,000 words)📅 Deadline: 24.06.2026
🔗 https://workshop26.content-moderation.org/Join us in Duisburg 🇩🇪🇪🇺 (near Düsseldorf)!
-
📢 CFP for our #MuC2026 workshop “Human-Centered #ContentModeration: Expertise, Context, and Evaluation” (30 Aug 2026)!
Submit:
📄 Position papers (2 pages)
✍️ Reflection essays (1,000 words)📅 Deadline: 24.06.2026
🔗 https://workshop26.content-moderation.org/Join us in Duisburg 🇩🇪🇪🇺 (near Düsseldorf)!
-
Ctrl-Alt-Speech: Message In A Bottleneck
https://fed.brid.gy/r/https://www.techdirt.com/2026/05/21/ctrl-alt-speech-message-in-a-bottleneck/
-
Ctrl-Alt-Speech: Message In A Bottleneck
https://fed.brid.gy/r/https://www.techdirt.com/2026/05/21/ctrl-alt-speech-message-in-a-bottleneck/
-
We Need A More Serious Discussion About Suicide And AI Chatbots
-
We Need A More Serious Discussion About Suicide And AI Chatbots
