Reaction to Meta's T&S about-turn, safety tech aquisition and Discord bot builders
Hello and welcome to the Everything in Moderation's Week in Review, your need-to-know news and analysis about platform policy, content moderation and internet regulation. It's written by me, Ben Whitelaw and supported by members like you.
Some people say its too late to say HNY but, in times like these, all rules go out of the window. So happy new year to all EiM subscribers, including folks from Teleperformance, Google, Nisos, ActiveFence, Global Counsel, eSafety Commission, Milltown Partners and others that signed up in recent weeks.
While many expected a pre-inauguration shuffling of public policy deckchairs, few people could've predicted the content of Meta's announcement on Tuesday. You'll find my take and some required reading in today’s Platform section.
Zuck, however, isn't the only one with a big announcement this week: I'm pleased to say that EiM is doing its first meetup event! It'll be in London at the end of the month and is in partnership with two stellar tech policy newsletters. You should come along if you're in the UK. More details below.
If your New Year's resolution is to be informed about online speech, take out an EiM membership for full access to its archive, including previous editions that I've linked to in today's newsletter (marked #64, #263 etc).
Get in touch if you have thoughts about today's round-up. Here's everything in moderation this week— BW
Join me to talk 2025 tech policy with two of my favourite newsletters
Want to figure out the big tech policy trends for 2025? Keen to meet other folks that are interested in online speech and internet safety? You're in luck.
I'm teaming up with Digital Politics' Mark Scott and Georgia Iacovou of Horrific/Terrific for an in-person discussion/drinks on Thursday 30th January between 7-9pm. Venue and agenda TBC but it'll be informal and fun.
We've got dozens of people already signed up — why not join them?
Policies
New and emerging internet policy and online speech regulation
Elon Musk’s proactive support of Germany’s far-right party over the last weeks has woken up European lawmakers and led to some movement on the ongoing European Commission investigation into X/Twitter (EiM #229). A letter from EU tech policy lead Henna Virkkunen (EiM #263) and justice chief Michael McGrath to elected officials said they were moving forward “energetically”. Whatever that means. Bloomberg has the full story.
If you weren’t tuning into to Vietnamese regulatory news over the festive break — and you'd be forgiven for doing so — you may have missed that online speech law, Decree 147, (EiM #186) came into force on Christmas Day. The law, which mandates user identification and content takedowns for social platforms, gives more power to the Vietnamese government to stamp out dissent and has been called “draconian” by Human Rights Watch.
Also in this section...
- Greece announces plans to decrease young people's internet use (Euronews)
- The Great Decentralization (Noema)
- The Pro-Eating-Disorder Internet Is Back (The Atlantic)
Products
Features, functionality and technology shaping online speech
WebPurify, one of the many automated text and multimedia moderation technologies on the market, has this week been acquired by IntouchCX, the customer support Business Process Outsourcing (BPO) company with large workforces in Asia and Latin America.
The move could be an early signal of two trends that I’m keeping an eye out for in 2025: 1) the acquisition and consolidation of smaller safety tech providers that weren't able to carve out significant market share and 2) BPOs seeking to grow their T&S business through tools and services as well as traditional headcount.
Related: many of you enjoyed this week’s T&S Insider newsletter in which Alice wrote about a new report charting the growing technology and tools market. Go have a read.
Also in this section...
- How to Build Governable Spaces for Online Communities (Media Economies Design Lab)
Platforms
Social networks and the application of content guidelines
The stilted delivery. The watch. The tech bro aesthetic. The 931 carefully chosen words. Over the last few days, much has been written about Meta's ‘more speech and fewer mistakes’ announcement. I expect we’ll be talking about the consequences of Mark Zuckerberg's video in five, perhaps ten, years from now.
There's a lot of good analysis about Meta the company (we'll be discussing it on this week's Ctrl-Alt-Speech) but, for brevity's sake, I've only included the commentary that looks at the online speech elements of the announcement below:
Replacing fact-checkers with community notes: Alexios Mantzarlis of Faked Up fame is well-versed on both fact-checking and the T&S world and writes a withering critique of Zuck’s justification, pointing out that Meta partners mainly flag “low-quality spammy clickbait that Meta platforms have commodified”. Scientist Tom Stafford’s assessment of Community Notes is also as comprehensive as I’ve seen.
Simplifying content policy: Along with the civic content news, this was arguably, the most shocking and political part of the announcement. Platformer's Casey Newton spoke to 10 current of former employees, who told him the changes would likely lead to substantial increases in hate speech. Civil society orgs echoed these worries and Meta employees are beginning to speak out. A real mess.
Reducing mistakes: The company have known for many years that automated mistakes have led to speech suppression and shadow banning (EiM #220 and others). Heck, Nick Clegg, Kaplan's predecessor, even trailed this in December (EiM #274). Civil society orgs, such as the Electronic Frontier Foundation, were quick to welcome this focus but, allied with the simplified content policy, it could indirectly see a lot more users face harm on the platform.
Alice Hunsberger
Bringing back civic content: Less has been made of the u-turn on its 2021 decision to scale back political content on Facebook, Instagram and Threads but it has big implications for users and publishers. There's some confusion around how it will work, though: TechCrunch noted that Facebook will replicate the political content controls that Instagram launched in February 2024 (EiM #235) although BBC reported that users will no longer be able to opt out completely.
Moving moderators to Texas: Facebook has had moderators in Texas for years, which we know because they brought a class-action suit against the company back in 2020 (EiM #64). But, as Kate Klonick points out in this excellent Lawfare panel discussion, this was the part of the announcement designed to signal a move away from the "Californian elite".
The sixth and bonus announcement — that Meta would work with President-elect Trump to push back against “governments around the world that are going after American companies and going to censor more” — was firmly pushed back on by the European Commission, which “absolutely refute[d] any claims of censorship”. Interestingly, the New York Times suggested that this is less about Zuckerberg's “cynical opportunism” and more about his “real enthusiasm” for aspects of MAGA conservatism. Hmmm.
Whatever his motivation — and, for me, this is the crux of it all — Zuck can expect other political leaders at his door asking for the same Trump treatment. Academic Joao C. Magalhães writes that:
“Authoritarians across the world are certainly salivating at the prospect of having an unprecedentedly sophisticated global communication system at their disposal.”
It’s rare for X/Twitter and Elon Musk to be playing second fiddle on the speech news stakes but, god bless him, he’s trying his best to tweet his way to some headlines. The Financial Times analysed his recent posts and found that his obsession with UK politics could be being fuelled by a series of right-wing accounts pushed to him by his own “For You” algorithm.
Also in this section...
- Instagram blocked teens from searching LGBTQ-related content for months (User Mag)
- The erasure of Luigi Mangione (Substack)
- The Video Game Industry Is Finally Getting Serious About Player Safety (Wired)
People
Those impacting the future of online safety and moderation
I’m not a big Discord user but lots of people are and many of them use MEE6, an all-in-one bot that streamlines community moderation and server management. It was created by two friends, Brendan Rius and Anis Belkacem, and is now used by over 22 million servers worldwide.
The bot has not been without controversy; it was the suspected vector of a crypto discord hack in 2022 and has been criticised for trying to sell out its loyal users. But I still got something from this (paid for) Inverse profile in which Rius and Belkacem talk about building the tool and moving to France to save on rent as they scaled. Building mod tools — as anyone who has done it knows — is hard.
Posts of note (non Meta-related edition)
Handpicked posts that caught my eye this week
- “This survey is available for all social media users, not just members of my instance, and not just Mastodon or fediverse users.” - Friend of EiM david wolfpaw is running anonymous (and timely) survey about transparency in mod decisions. Take a second to fill it out.
- "Go beyond surface-level symptoms. Identify true root causes, address systemic issues, and reduce the risk of repeat incidents." - Former head of Twitter T&S Del Harvey shares the 7 Rules of Causation, part of an upcoming Root Cause Analysis guide from the Trust & Safety toolkit.
- "This year's theme is "Moderating Third Spaces," focusing on the transformative power of online moderation in building and restoring community through digital third spaces." - All Things in Moderation, the virtual conference I took part in two years ago and hope to join again this year, is accepting contributors and papers. Go apply!
Member discussion