Commenters on YouTube now hit a pre-publication warning when their draft text looks likely to offend, giving them a chance to edit instead of having the comment removed later.
Creators gain stronger built-in comment filters, reducing how much toxic or spammy text reaches their videos and how much time they spend moderating manually.
Second-order effects
The nudge model creates the infrastructure for harder enforcement: once YouTube can flag a comment before it posts, the same signals support the removal notices and temporary commenting bans it adopted two years later.
Competing video platforms face pressure to match pre-post intervention, since creators weigh moderation burden — not just revenue splits — when choosing where to publish.
Third-order effects
If the pattern holds, comment moderation across major platforms shifts structurally from reactive takedowns to algorithmic gating at the input box, with the platform's classifier effectively deciding what speech is worth attempting.
The risk side is visible in YouTube's own swearing rules, which it walked back after its November 2022 limits proved stricter than intended — automated behavioral policies at scale tend to require public recalibration, making policy reversals a recurring feature rather than an exception.
The trend: Platform moderation is migrating upstream from post-hoc removal to pre-publication AI nudges, with each intervention layer building toward automated enforcement of user behavior.
“When tech companies develop policies designed to manage hate and harassment, they owe it to the public to do evidence-based governance” says Nathan Matias, @natematias assistant professor of communication. By @adamndsmith via @Independent https://www.independent.co.uk/ ...
@oliviasolon
Olivia Solon on x
Huh. So Facebook did eventually listen to all the researchers who have been highlighting this issue for years https://www.independent.co.uk/ ...
@hypervisible
@hypervisible on x
“Before a comment is posted, a pop-up will appear with a reminder to ‘keep comments respectful.’” 😶 https://www.engadget.com/...
@theemilydbaker
@theemilydbaker on x
I wonder how much of our good hearted cursey words are going to get caught up in here. Can I apply for ‘fuckery’ to not be deemed offensive as a matter of fact @YouTubeCreators? #lawnerdsunite https://twitter.com/...
@antoniocasilli
Casilli on x
In your opinion, how much human content moderators will be needed to provide the illusion that this “new product” inviting people “to reflect before posting” offensive comments qualifies as “automatic filtering”? https://twitter.com/...
@tyadkisson
Tyler Adkisson on x
YouTube is introducing new features to help combat hate speech and make the platform more inclusive. One feature: a new “comment reminder” prompt that warns users when their post may be offensive to others, and gives the option to reflect before posting. https://blog.youtube/... …
@rkyncl
Robert Kyncl on x
We are dedicated to making sure our products & policies help @YouTube creators from all communities thrive. Check out this update on how @YouTube is implementing changes to better help the platform work for all: https://blog.youtube/...
@susanwojcicki
Susan Wojcicki on x
At YouTube, we've been looking closely at how our policies and products are working for everyone, and specifically for the Black community. Today we're sharing an update: https://blog.youtube/...
@akilahobviously
Akilah Hughes on x
I wonder which Black YouTube creators they consulted with and how much they didn't pay them. 🥱 https://twitter.com/...
@youtubecreators
@youtubecreators on x
Learn more about our ongoing efforts to make YouTube a more inclusive platform, including how we're using your feedback to improve the nature of comments and better serve creators. https://blog.youtube/...
@edbott
Ed Bott on x
Someone at YouTube thought this was a good way to deal with its toxic comments problem https://twitter.com/...
@sarafischer
Sara Fischer on x
NEW: YouTube making changes to limit hate speech, boost inclusion - Twitter, Facebook announced similar efforts around hate speech/inclusion this week -Simultaneous announcements suggest tech firms may have waited to introduce changes until after election https://www.axios.com/..…
@sullydish
Andrew Sullivan on x
I kept waiting in this piece for any voice that might speak up for race-neutral hate speech bans. But nah. CRT only in the WaPo. https://www.washingtonpost.com/ ...