Under-18 chatbot ban, OpenAI's cynical transparency and more researcher access
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.
Week in Review isn’t an AI safety newsletter but this week, it might as well be. From OpenAI’s push for transparency to Character.ai’s under-18 ban and a former insider’s explosive op-ed, the debate about what “safe” AI looks like — and who gets to define it — has dominated the discussion.
Which begs the question: Would you like more dedicated coverage or AI safety and governance from EiM? What are you missing from the deluge of AI analysis and hype already out there? Get in touch and tell me what you’d find most useful.
Welcome to new subscribers from Yubo, Roblox, GamerSafer, CloudKitchens, Bytedance, the Integrity Institute and other safety-minded folks from around the world. This is your Week in Review — BW
Introducing Trust Issues — Insights from the People Who Keep the Internet Safe, the new podcast from WebPurify, an IntouchCX company.
Each episode offers candid conversations with the people defining and influencing what’s allowed online, and what’s not – from product leaders to policy makers.
It's hosted by WebPurify's Head of Trust & Safety EMEA, Ailís Daly, a trained barrister-at-law, and the former Global Head of Violence & Aggression Policies at TikTok.
Episodes now available:
Ep 1: Fraud survivor advocate Cecilie Fjellhøy on rebuilding trust online
Ep 2: Google's Abishek Roy on teaching users to BeScam Ready
Ep 3: Dr Tracy Elizabeth on Instagram's PG13 content classification rating
These conversations cut through the noise to reveal how Trust & Safety really works. Listen and subscribe via Spotify and YouTube.
Policies
New and emerging internet policy and online speech regulation
Meta and Google executives this week faced fresh questioning from Senate Republicans over alleged Biden administration pressure to suppress COVID-19 and election content. Both companies acknowledged outreach from government officials with Meta’s Neil Potts saying the company believed that "pressure was wrong”. The hearing — led by Senator Ted Cruz and dubbed “Shut Your App, Part II” — also featured testimony from FIRE and Public Knowledge. Tech Policy Press has a very helpful transcript.
Last week’s Week in Review included news of the EU’s latest Digital Services Act ruling against TikTok and Meta (EiM #311) regarding researcher data access. This Science.org piece brings the investigation to life, with stories of researchers waiting more than a month to be rejected by platforms or, perhaps worse, receiving partial data sets. With new rules rolling out for researchers to gain previously unobtainable platform data, I expect more VLOPs will come unstuck.