Apple's 'surprising' T&S move, another teen ban and 63,000 kids target the Kremlin
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 paid members like you.
In recent years, there has been a growing call — particularly from T&S leaders at mid-sized and smaller companies — for operating-system owners to take a stronger role in online safety. This week, at Apple’s developer conference, that became a little more real (see: Platforms) and is arguably the biggest story of the week.
It’s not the only story in town, though. Canada has made a belated but significant attempt at online regulation, UK politics is threatening to turn child safety into another culture-war football (see: Policies), and Russian gamers have proved that, sometimes, the pen is mightier than the ban (see: People).
Last week, I promised to write something about Meta's new defaults (EiM #338). Time got the better of me — at the moment, EiM is being squeezed into the equivalent of one day a week — which is precisely why membership matters. By supporting financially, I have more time to produce independent, properly considered coverage of the T&S industry at a moment when it’s most needed.
Big welcome to new subscribers from Publiq, New York University, Snapchat, TaskUs, ByteDance, The Telegraph, Noosphr, Counterhate, Google and elsewhere. Hit reply, say hello, wherever you are in the world.
Here's your Week in Review — BW
If you moderate, or if you care about building safe, thriving digital spaces, this year's All Things in Moderation conference is for you.
Taking place virtually on June 25-26, ATIM is a must-attend event for moderators, community managers, researchers, policy-makers and technologists who want to learn from each other, network and collaborate.
The agenda looks great — including folks from BlackSky, commercial moderators from India and conspiracy theory and manosphere experts — and, if you needed a push to register, EiM subscribers can use the code 'EIM' to get discounted tickets. See you there!
Policies
New and emerging internet policy and online speech regulation
Canada has become the latest country to decide that online safety policy should start with children. The federal government’s Safe Social Media Act would require regulated social media services to prevent under-16s from having accounts, while creating a new Digital Safety Commission of Canada to set standards, assess compliance and manage complaints. The Globe and Mail has more.
Down under differential: Unlike Australia’s social media ban — which is already running into implementation trouble (EiM #331) — Canada’s approach allows platforms to apply for an exemption if they have “adequate safeguards” for children. In other words, guilty until proven innocent — which is going to be incredibly hard to evaluate.
As UK politicians appealed for calm after violence reportedly co-ordinated via social media in Belfast, Prime Minister Sir Keir Starmer continued to bungle the politics — and product reality — of online safety:
- At London Tech Week, he called on Apple and Google to block nude images on children’s devices within three months, only for Apple to announce a similar feature at its developer conference later the same day (see: Platforms)
- According to reports, No 10 is ploughing ahead with an under-16 social media ban, with an announcement possible as soon as next week as Labour stares down a crucial by-election.
Frankly, it all has the feel of online safety by media release. Which, as I’ve written before, is rarely a good thing.
Ben Whitelaw
Also in this section...
- Top Trump official slams Germany (again) over verdict for man who called Merz a liar (Politico)
- The DSA’s Sleeping Giant is Waking Up (Tech Policy Press)
- State of Play of DSA Dispute Settlement: Meaningful Redress, Uneven Results (DSA Observatory)
Products
Features, functionality and technology shaping online speech
Mythos, the “too dangerous” Anthropic model which was given to a select group of companies for testing, has been made safe for general use and released as Fable 5 (EiM #333). In practice, that means that queries related to cybersecurity, biology and chemistry will default to a less capable model. And it seems Anthropic is being cautious: the announcement notes that this model downgrade feature could happen in 5% of sessions — which may not go down well with its corporate subscribers, which account for 80% of its revenue.
Ben Whitelaw
Also in this section...
- Coop 1.0: World’s First Free, Open Source Child Safety Infrastructure for Every Platform (ROOST)
- We must not grant AI agents legal personhood (Financial Times)
- Due Process: Why Fair Moderation Builds Healthier Communities (Prosocial Design Network)
💡 Become an individual member and get access to the whole EiM archive, including the full back catalogue of Alice Hunsberger's T&S Insider.
💸 Send a tip whenever you particularly enjoyed an edition or shared a link you read in EiM with a colleague or friend.
📎 Urge your employer to take out organisational access so your whole team can benefit from ongoing access to all parts of EiM!
Platforms
Social networks and the application of content guidelines
Apple put Trust & Safety front and centre at its WWDC developer conference this week as it launched a host of child safety measures that make it a more important component in the online safety stack. It’s worth watching the ten-minute segment — featuring longstanding employees Raja Bose (Director of Trust, Safety & Values) and Ann Thai (Senior Director, Apple Marketplace Platforms & Technologies) - in full. But if you’re short on time, here’s what’s coming:
- An Ask To Browse feature, which allows children to ask their parents to access specific websites.
- Images sent via Message or live FaceTime calls that contain gore or violent content will be blurred (in addition to nudity).
- More detailed Screen Time allowances across Entertainment, Games and Social Media and scheduling mechanism with recommendations from the American Academy of Pediatrics and other experts.
Advantage Apple? Industry analyst Benedict Evans called it “pretty surprising and striking” and potentially “a new marketing story on a par with privacy”. The focus certainly differed from Google’s recent own developer conference, which contained a Chrome app download feature and device theft detection but otherwise had little in the way of child safety updates. For my part, as I mentioned a few weeks back, I’d like to see more transparency about its safety efforts, especially as they ramp up.
Ben Whitelaw
Following the California and New Mexico verdicts (EiM xx), many — including myself — are wondering which lawsuits are coming down the track. The BBC has done a good roundup of four important cases against Roblox, Meta (x2) and Instagram, YouTube, Snapchat and TikTok to watch.
Also in this section...
- AI ‘content creators’ are getting harder to spot (The Verge)
- Social media on trial: Four important cases to watch (BBC)
People
Those impacting the future of online safety and moderation
Most weeks, EiM's People section is reserved for researchers, platform workers or activists shaping the future of online safety. This week, it belongs to 63,000 Russian children who have achieved what many opposition politicians, civil society groups and sanctioned tech companies have not: making the Kremlin look silly.
According to TNW, Russia has reportedly unblocked Roblox after a wave of complaints from children over its December ban on the platform. The block — introduced back in December last year — was dressed up in familiar child-protection language, with authorities citing extremism, LGBT “propaganda” and inappropriate content.
But Russian authorities presumably didn’t expect tens of thousands of letters, many from 8- to 16-year-olds, asking for a reversal and threatening to leave the country if Roblox stayed banned. I’m curious as to who co-ordinated the letters and what their motive is (if you know more, get in touch). But I love the result.
Posts of note
Handpicked posts that caught my eye this week
- “It is a useful reminder that young people are already using AI and digital tools to learn, create, revise, translate, problem solve and prepare for work.” - Save The Children’s Jeffrey DeMarco reflects on a new Google report on teen use of AI.
- “How do you attribute a hiring scam to North Korea when the interviewers are in Colombia, Nigeria and the Philippines, the LinkedIn accounts are rented in Bangladesh, and the boss never shows his face?” - Craig Silverman, of Indicator Media, with a banger of an investigation.
- “We got some hard facts and hard stats on Meta's demonetization of repeat disinformers. A quick spoiler: it's not looking good.” Victoire Rio of What To Fix with a report on why mis- and disinformation remains a profitable game.
Member discussion