Simon Munzert
@simonsaysnothin.bsky.social
Professor of Data Science and Public Policy | Hertie School Data Science Lab | Elections, Public Opinion, Data
This week, the 2025 edition of the Lancet Countdown Report on Health and Climate Change was released — and I was able to contribute a small part again (look for indicator 5.2 in our version of the #ShowYourStripes chart, that’s me!). 1/4
October 31, 2025 at 8:43 AM
This week, the 2025 edition of the Lancet Countdown Report on Health and Climate Change was released — and I was able to contribute a small part again (look for indicator 5.2 in our version of the #ShowYourStripes chart, that’s me!). 1/4
Main finding #2: Group markers drive over-moderation. Words like "muslim", "gay", or "jews" make mis-classifying non-hate speech as hate speech more likely.
May 12, 2025 at 8:25 PM
Main finding #2: Group markers drive over-moderation. Words like "muslim", "gay", or "jews" make mis-classifying non-hate speech as hate speech more likely.
Main finding #1: Performance varies wildly across moderation services, datasets, and metrics, but some of the failure rates are astonishing (FPR & FNR > 75% on balanced samples for some implicit and explicit speech!).
May 12, 2025 at 8:25 PM
Main finding #1: Performance varies wildly across moderation services, datasets, and metrics, but some of the failure rates are astonishing (FPR & FNR > 75% on balanced samples for some implicit and explicit speech!).
Our paper investigates how commercial content moderation APIs handle group-targeted hate speech. This is a black-box audit (we have no idea how the models look exactly) of 5 major APIs with five million queries based on four datasets.
May 12, 2025 at 8:25 PM
Our paper investigates how commercial content moderation APIs handle group-targeted hate speech. This is a black-box audit (we have no idea how the models look exactly) of 5 major APIs with five million queries based on four datasets.
Have you ever used a content moderation API, such as Perspective API or OpenAI's Moderation API, for your research or to inform moderation decisions? Well, they might not have given you what you think they would.
May 12, 2025 at 8:25 PM
Have you ever used a content moderation API, such as Perspective API or OpenAI's Moderation API, for your research or to inform moderation decisions? Well, they might not have given you what you think they would.