Child safety deadline looms, Eurosky lifts off and new sextortion documentary

Hello and welcome to 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.

With the UK’s child safety codes coming into play next week, the giant platform poker game over age assurance is entering its final round. A host of platforms have shown their hands but the rest? Still holding their cards close. The stakes are high so we’ll see who folds and who tries to bluff their way through.

If you're heading to San Francisco for TrustCon, a) download the latest episode of Ctrl-Alt-Speech so you can listen to the dulcet tones of Mike and I en route to the west coast and b) don't miss the second live recording of the podcast on Wednesday afternoon.

Welcome to new subscribers from the frontlines at Helpshift, Patreon, Aura, Spotify, Salesforce and others — get in touch to say hi and share what you're working on. This is your Week in Review — BW

Policies

New and emerging internet policy and online speech regulation

With new Ofcom child safety codes coming into force next week (25th July), all the talk is about the implementation of age assurance measures for platforms with a UK presence. 

This week, we saw Bluesky expand on its plans to verify users’ ages (using Epic Games’ Kids Web Services, if you’re wondering) while Reddit will use Persona which, I noted, already works with OpenAI to screen users. Both platforms follow Pornhub’s announcement two weeks back (EiM #295) and will no doubt be followed by other platforms this week. I expect some to be more forthcoming than others with their plans.

Media watch: News coverage of the codes has ramped up and I was very interested in the interview given by Ofcom CEO Melanie Dawes to the BBC, in which she called it a “really big moment” for children’s safety. Putting aside the content of the codes for a second, I was impressed by Dawes’ understanding of safety trade-offs and her pushing back against some ill-informed questions by one of the BBC’s highest-profile presenters. More analysis on this story in this week’s Ctrl-Alt-Speech. 

It’s worth noting that this is happening at the same time as new rules come into play in Ireland, which — according to The Gist’s Simon McGarr — are “manifestly insane”. The EU also outlined its approach for child safety this week, which includes a bloc-wide age verification app.

In not-unrelated news, The Wikimedia Foundation has created a handy checklist designed to help policymakers assess whether proposed regulation would affect Wikipedia’s core operating model — or indeed any platforms that provide information in the public interest. The Wikipedia Test’s seven questions cover access to information, free expression and privacy & safety with some helpful context for each. One wonders whether someone in the Ofcom office is furiously completing the test right now…

Also in this section...

• Age Verification Laws: Are the New Social Networks Different, Or Not At All?(Connected Places)
• Brussels stalls probe into Elon Musk’s X amid US trade talks (Financial Times)

Products

Features, functionality and technology shaping online speech

Eurosky, a Europe-built and funded moderation service, launched this week aimed at reducing dependency on US tech platforms. Using the AT Protocol developed by Bluesky, it seeks to reduce the cost of moderation operations and help platforms adhere to government regulations (read EU ones) more easily. Linked to both ROOST (EiM #281) and the Free Our Feeds campaign (EiM #277), Eurosky is a big digital sovereignty bet — and one I expect to return to in EiM in future.

What's in a name?: Could the brand be, erm, better? Certainly. Currently searching for Eurosky gets you a private jet company, a Liberian cargo ship and a Polish TV company among others things (The website is eurosky.social fwiw).

Also in this section...

• Resolver Unveils Powerful New Service for Identifying Unknown CSAM (Resolver)

Platforms

Social networks and the application of content guidelines

Hugging Face — a popular platform for people to share sharing AI models — has been inundated with LLMs used to generate Non Consensual Intimate Imagery (NCII), posing a platform governance problem for the French-founded startup. 

404 Media reports that around AI image generation 5,000 models were re-uploaded there after being removed from Civitai, another community-drive AI model platform. However, that these models don’t technically violate Hugging Face’s policies because the user bears responsibility for their use. Not only that but they’ve been given anodyne names — like ‘test model’ so taking them down isn’t straightforward. 

Where the real power lies: The reason why all this happened in the first place is because payment processors kicked up a fuss — understandably — about the proliferation of adult content on Civitai. That, allied with the pressure being put on gaming store Steam this to clean up its act, reminds us that financial institutions are a bigger part of the T&S ecosystem than we sometimes think. 

Also in this section...

• Grok Chatbot Mirrored X Users’ ‘Extremist Views’ in Antisemitic Posts, xAI Says (New York Times)
• Grok's new porn companion is rated for kids 12+ in the App Store (Platformer) 

People

Those impacting the future of online safety and moderation

Although most/I know him as one half of UK hip duo Rizzle Kicks (which released the 2012 chart hit Mama Do The Hump — check it out) , Jordan Stephens does a solid turn as a director and has turned his attention to sextortion for his latest film. (UK readers can watch it here).

In an interview given to The Times to promote it, Stephens explains how he set up a honeytrap to find scammers operating out of a barber shop in Nigeria. Not only does he track them down but he interviews them about why they do it and the lucrative rewards if they are successful. I haven’t watched the full film yet but it sounds right up my street.

There's a strong irony — with the widespread happening in the UK and elsewhere about age verification — that Stephens believes “you should have to at least provide ID to make a social media account”.  There are trade-offs with that, as we know, but luckily, there are no downsides with this interesting documentary.

Posts of note (TrustCon edition)

Handpicked posts that caught my eye this week

• "TrustCon is one of my absolute favorite events for Trust & Safety professionals to share stories, meet and learn. " - Henriette Cramer has a panel on the intersection of safety and safety that looks great.
• "This is a pivotal moment for the trust and safety profession—as online harms grow more complex, and expectations around transparency, well-being, and accountability continue to rise, our collective work has never been more critical." - Ljubiša Velikić on why this TrustCon stands out.
• "There's something energizing about being around people who care deeply about the same problems as we do. Building trust online isn't easy, but the community working on it is quite amazing." - If you've been to TrustCon, you'll know Andy Yang is right.