# Digital Guardians: Can GPT-4, Perspective API, and Moderation API reliably detect hate speech in reader comments of German online newspapers?

arXiv: https://arxiv.org/abs/2501.01256 · HTML: https://mlsanity.com/paper/2501.01256/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 35 references that are in the map.
- First posted on arXiv 2 Jan 2025.

## Builds on (most cited first, 16 of 16)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation](https://mlsanity.com/paper/2010.16061/) (2020)
- [A Holistic Approach to Undesired Content Detection in the Real World](https://mlsanity.com/paper/2208.03274/) (2022)
- [A New Generation of Perspective API: Efficient Multilingual Character-level Transformers](https://mlsanity.com/paper/2202.11176/) (2022)
- [Charformer: Fast Character Transformers via Gradient-based Subword Tokenization](https://mlsanity.com/paper/2106.12672/) (2021)
- [German's Next Language Model](https://mlsanity.com/paper/2010.10906/) (2020)
- [Deceiving Google's Perspective API Built for Detecting Toxic Comments](https://mlsanity.com/paper/1702.08138/) (2017)
- [Detecting Hate Speech with GPT-3](https://mlsanity.com/paper/2103.12407/) (2021)
- ["HOT" ChatGPT: The promise of ChatGPT in detecting and discriminating hateful, offensive, and toxic comments on social media](https://mlsanity.com/paper/2304.10619/) (2023)
- [A systematic review of Hate Speech automatic detection using Natural Language Processing](https://mlsanity.com/paper/2106.00742/) (2021)
- [Toxic Bias: Perspective API Misreads German as More Toxic](https://mlsanity.com/paper/2312.12651/) (2023)
- [An Investigation of Large Language Models for Real-World Hate Speech Detection](https://mlsanity.com/paper/2401.03346/) (2024)
- [Close to Human-Level Agreement: Tracing Journeys of Violent Speech in Incel Posts with GPT-4-Enhanced Annotations](https://mlsanity.com/paper/2401.02001/) (2024)
- [A Comprehensive Study on NLP Data Augmentation for Hate Speech Detection: Legacy Methods, BERT, and LLMs](https://mlsanity.com/paper/2404.00303/) (2024)

## Built on by (most cited first, 3 of 3)

- [Detecting and Filtering Unsafe Training Data via Data Attribution with Denoised Representation](https://mlsanity.com/paper/2502.11411/) (2025)
- [SpatialJB: How Text Distribution Art Becomes the "Jailbreak Key" for LLM Guardrails](https://mlsanity.com/paper/2601.09321/) (2026)
- [Decoding Islamophobic Discourse: Using LLMs to Identify Tropes and Semi-Coded Hate Speech](https://mlsanity.com/paper/2503.18273/) (2025)

## Abstract (arXiv)

In recent years, toxic content and hate speech have become widespread phenomena on the internet. Moderators of online newspapers and forums are now required, partly due to legal regulations, to carefully review and, if necessary, delete reader comments. This is a labor-intensive process. Some providers of large language models already offer solutions for automated hate speech detection or the identification of toxic content. These include GPT-4o from OpenAI, Jigsaw's (Google) Perspective API, and OpenAI's Moderation API. Based on the selected German test dataset HOCON34k, which was specifically created for developing tools to detect hate speech in reader comments of online newspapers, these solutions are compared with each other and against the HOCON34k baseline. The test dataset contains 1,592 annotated text samples. For GPT-4o, three different promptings are used, employing a Zero-Shot, One-Shot, and Few-Shot approach. The results of the experiments demonstrate that GPT-4o outperforms both the Perspective API and the Moderation API, and exceeds the HOCON34k baseline by approximately 5 percentage points, as measured by a combined metric of MCC and F2-score.