Bye Bye Perspective API: Lessons for Measurement Infrastructure in NLP, CSS and LLM Evaluation

Perspective API, the de facto standard for toxicity measurement, will shut down at the end of 2026, exposing researchers' dependence on a tool they did not control, according to a paper by David Hartmann and colleagues. Surveying 241 papers, the authors found that reliance on Perspective led to unsupported claims, shifting results from silent model retraining, and unexplained disparities, which were amplified throughout the LLM lifecycle. They release Perspective scores for 5.9 million text snippets from 77 datasets and specify ten requirements for field-owned measurement infrastructure.

Computer Science Computation and Language Submitted on 28 Apr 2026 v1 https://arxiv.org/abs/2604.25580v1 , last revised 31 Aug 2026 this version, v2 Title:Bye Bye Perspective API: Lessons for Building and Governing Measurement Infrastructure View PDF /pdf/2604.25580 HTML experimental https://arxiv.org/html/2604.25580v2 Abstract:Perspective API closes at the end of 2026, removing the de facto standard for toxicity measurement and exposing researchers' dependence on a tool they did not control. Drawing on this case, we argue that a research field must build and govern its own measurement infrastructure rather than borrow it. Surveying 241 papers that use or study Perspective, we show what depending on it cost the research community: claims reaching past what the tool could support, results that shifted when its model was silently retrained, and disparities researchers could measure but not explain. These failures were amplified throughout the LLM lifecycle, where Perspective supplied the labels, filtered the corpora, and graded the systems trained on each, rewarding errors rather than catching them. To keep the instrument open to study after shutdown, we release Perspective scores for 5.9 million text snippets from 77 datasets. We further specify ten requirements for measurement infrastructure a field owns, and argue that what blocks such infrastructure is not technical capability but the value the field places on infrastructure work. Submission history From: David Hartmann view email /show-email/1cbf4cb3/2604.25580 Tue, 28 Apr 2026 12:49:54 UTC 394 KB v1 /abs/2604.25580v1 v2 Mon, 31 Aug 2026 13:45:30 UTC 478 KB References & Citations Loading... Bibliographic and Citation Tools Bibliographic Explorer What is the Explorer? https://info.arxiv.org/labs/showcase.html arxiv-bibliographic-explorer Connected Papers What is Connected Papers? https://www.connectedpapers.com/about Litmaps What is Litmaps? https://www.litmaps.co/ scite Smart Citations What are Smart Citations? https://www.scite.ai/ Code, Data and Media Associated with this Article alphaXiv What is alphaXiv? https://alphaxiv.org/ CatalyzeX Code Finder for Papers What is CatalyzeX? https://www.catalyzex.com DagsHub What is DagsHub? https://dagshub.com/ Gotit.pub What is GotitPub? http://gotit.pub/faq Hugging Face What is Huggingface? https://huggingface.co/huggingface ScienceCast What is ScienceCast? https://sciencecast.org/welcome Demos Recommenders and Search Tools Influence Flower What are Influence Flowers? https://influencemap.cmlab.dev/ CORE Recommender What is CORE? https://core.ac.uk/services/recommender arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs https://info.arxiv.org/labs/index.html .