Trust & Safety

by hailey @hailey.at

Posts about trust and safety on social media and the internet — content moderation, platform policy, online safety regulation, disinformation, and related software and tooling.

Open and pin on Bluesky Share on Bluesky
0
Likes
+0
Growth per month

In the feed right now

Live from Bluesky
Aida @aidap.bsky.social · 2 days ago
Bluesky's Head of Trust and Safety handed his block list to a bot. 1,271 people blocked in eleven days. He says it never looks at who you are. Blocks are public, so I checked. Trans people were blocked at 2.5× the odds of everyone else, for posts of the same harshness. 🧵
Title card reading "Who the autoblocker blocked". Three figures: 1,271 accounts blocked in eleven days, against twelve in the two years before. 46% of the readable bulk-loaded accounts had never replied to, quoted or mentioned him this year. 2.5× the odds of being blocked for a trans-identified account, against others who wrote equally harsh posts. Below, a quote from Aaron Rodericks, 1 Oct 2026: "These do not factor in your bio or your identity. I have zero clue who you are."
Replies45 Reposts1098 Likes2728
Boo!-per Lund @cooperlund.online · 2 days ago
This isn’t an AI story, it’s a Trust and Safety story and Meta completely dropped the ball because giving someone the ability to pay for something that denotes trust and status is a Bad Idea.

You may have seen a viral picture of actress Mia Goth seemingly referencing her recent divorce filing. It was posted by an Instagram page w/ 1M+ followers and a blue checkmark. It was AI. Here's a story about multiple systems of verification/trust failing: www.theverge.com/ai-artificia...
“Mia Goth does not have ownership of, nor does she engage in social media on any platform. We have filed multiple complaints with Instagram to remove the verified fan account which continues to post counterfeit, AI images of Mia Goth,” Christine Tripicchio, a representative for Goth, told The Verge in an email. “The fact that this fan account meets Instagram’s criteria to qualify for a ‘blue check’ is disconcerting on a fundamental level. Beyond that, it is incomprehensible that Instagram evidently prioritizes the dissemination of forged and manipulated images over the protection and integrity of a public figure’s likeness.”
An AI photo of the actress Mia Goth wearing a shirt that reads "MY EX IS MY BIGGEST FAN."
Replies4 Reposts46 Likes336
Josiah Mortimer @josiahmortimer.bsky.social · 2 days ago
A bleak question I'd be keen to hear from UK lawyers about. Will the US' public execution - if it goes ahead - be blocked in the UK, under the Online Safety Act (unless the platforms showing it have age verification)? Or some other law? Or is this situation so mad/unique that we don't know?
Replies10 Reposts65 Likes200
Mitch Goldstein @mgoldstein.nyc · 2 days ago
We've been past the point of no return for a while now, Meta's business interests (sell more AI tools, encourage more AI in-app, pay for a blue check and Meta plus subscription) are in direct conflict with its Trust and Safety interests.

This isn’t an AI story, it’s a Trust and Safety story and Meta completely dropped the ball because giving someone the ability to pay for something that denotes trust and status is a Bad Idea.
Replies4 Reposts18 Likes147
aly @aly.codes · 1 day ago
you guys have been given an *unbelievable* amount of leeway precisely because you are targeting the head of trust and safety. you rightfully should be banned for this. Aaron did not sign up for personally targeting harassment by working at Bluesky.

Not sure if you’re joking so just in case: If directed at some random regular user/person on the street, sure, agreed If directed at people of large influence over relevant populations, absolutely warranted. Aaron & the like need to be able to handle it. This is the good kind of free speech
Replies4 Reposts0 Likes72
Aimee Cozza Illustration @aimeecozza.com · 3 days ago
Cool art a post back if you have non-sexual nudity set to HIDE and not WARN. It's set to this by default, you can change it under bsky.app/moderation or Settings > Moderation settings. I have mine set to SHOW bc idgaf but you should set yours on these 2 settings to WARN! You may be missing great art
Bluesky content filters: Sexually suggestive and non-sexual nudity options circled in red, arrow pointing to "warn" instead.
Replies1 Reposts55 Likes89

Hidden, might be NSFW

James D'Amato @oneshotrpg.bsky.social · about 23 hours ago
Odd choice for the head of "trust and safety" to document Blue Sky's preference for punishing trans people, but I can say it absolutely impacts my trust so I guess there's that.

Bluesky's Head of Trust and Safety handed his block list to a bot. 1,271 people blocked in eleven days. He says it never looks at who you are. Blocks are public, so I checked. Trans people were blocked at 2.5× the odds of everyone else, for posts of the same harshness. 🧵
Title card reading "Who the autoblocker blocked". Three figures: 1,271 accounts blocked in eleven days, against twelve in the two years before. 46% of the readable bulk-loaded accounts had never replied to, quoted or mentioned him this year. 2.5× the odds of being blocked for a trans-identified account, against others who wrote equally harsh posts. Below, a quote from Aaron Rodericks, 1 Oct 2026: "These do not factor in your bio or your identity. I have zero clue who you are."
Replies2 Reposts6 Likes23
Elizabeth Minkel @elizabethminkel.bsky.social · 1 day ago
Folks who look at fic often treat the big archives as wholly representative, but there are so many subcultures within our subculture. @bookshop.bsky.social reports a fascinating story about rigid content moderation, self-publishing, and what happens when the wide-open spaces of fic are constrained:
Pride and Prejudice in the Jane Austen Fanfic Community

fansplaining.com

Pride and Prejudice in the Jane Austen Fanfic Community

A 30-year-old moderating decision still defines the Austen fic fandom’s identity.
Replies0 Reposts6 Likes24
Keep reading on Bluesky