Reuters Digital News Report: For the first time, social media and video networks are, on average across the markets covered, more popular than both TV and owned news websites and apps as sources of news.
reutersinstitute.politics.ox.ac.uk/digital-news...
There's a major gap in content safety tooling: classifiers typically only score complete text. When you're working with generative AI, "complete text" means the user already saw it. That's too late.
So we built a streaming classifier that we're releasing today! Here's what we did and why.
🧵...
Katie Harbath
Does anyone else think that this monstrosity on the South Lawn is right out of the dystopian 1985 in Back to the Future? Biff even looks like Trump.
@katieharbath.bsky.social and I found that we have many mutual subscribers to our newsletters (@eimdotco.bsky.social and Anchor Change) and that they value how differently we cover tech policy and T&S.
So we found a way to sign up for both in one go. Fill yer boots!
trustfnd.com/katieharbath...
In Meta’s Q1 earnings call, they focused on how AI is making our feeds more personal.
It’s making me think about how we manage the tradeoffs between what's optimized for each of us individually might not be automatically good for us collectively.
anchorchange.substack.com/p/ai-optimiz...
Missed this on Friday. Anthropic dropped an update on Claude's midterm preparedness including work on political bias evaluations, policy enforcement mechanisms, and influence operation resistance testing. www.anthropic.com/news/electio...
Everyone's talking about AI in campaigns, but it’s hard to measure.
That’s why Amanda Elliott, @kylewilsontharp.bsky.social, @ericwilson.bsky.social and myself are running a survey on how political pros are using AI.
If you work in campaigns and politics we need your input!
gforms.app/j5nTd6r
Oh how interesting! Here they are:
zentropi.ai/labelers/cc7...
zentropi.ai/labelers/f51...
In 2017, it took us months to define “political ad” at Facebook. Recently, I built two political content classifiers in an afternoon using Zentropi AI created by @dwillner.bsky.social and @samidh.bsky.social
Why AI content moderation is good, actually — in this week’s newsletter 👇
Katie Harbath
Katie Harbath
Katie Harbath
Katie Harbath
Katie Harbath
Katie Harbath
Samidh
Thrilled to publish my latest article
BAN COOKIE BANNERS: A CASE STUDY IN TECH REGULATION
with the Harv. J. of Law & Tech!
A cri de cœur to end this worse-than-useless regulatory compliance regime and let something that actually protects users take its place.
papers.ssrn.com/sol3/papers....
1/
Our 2026 report finds news audiences around the world reacting with growing unease to successive episodes of political, economic, and technological turbulence. Assumptions about the way the world work...
