Ethical and social risks of harm from Language Models
Work-level record. Its arXiv versions and any conference or journal version are listed as sources, each with its own date. A citation to the work does not identify which version the citing author read.
- Issued by
- No issuing record yet. A missing record is a gap in coverage.
- Authors credited
- Geoffrey Irving · Lisa Anne Hendricks · Mia Glaese
- Dates
- Published 2021-12-08 authorship evidence authorship evidence authorship evidence
- Exact source
- Ethical and social risks of harm from Language Models arXiv · published 2021-12-08 · arXiv abstract page, retrieved for the candidate pack on 2026-10-28 (SHA-256 fdd8d443b655364854a6b3fe1ad968935e001ff4278b7724d2284bb0c1148750). Submission history as arXiv records it (UTC): v1 8 December 2021. The page serves the latest version; each version is a separate expression, and only the one a record names was read.
- Ethical and social risks of harm from Language Models arXiv · published 2021-12-08 · arXiv:2112.04359v1, submitted 8 December 2021 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 515dd05c875f2d329cc035f081fcb1982d6a38884fa3476cc08ba8d6d850a2c1). The author list is the byline as printed on the first page.
Appears in
No collection holds a record about this document yet.
Versions
No other versions of this document are recorded. The evidence applies to the source version listed above. Other versions, such as a preprint, working paper or journal publication, are not identified or substituted.
Issue and authorship records
- Author: Geoffrey IrvingGeoffrey Irving is a credited author of Ethical and social risks of harm from Language Models.2021-12-08 · publication · Stated in source · Supported Inspect claim & evidence ↗Geoffrey Irving: person page →
- Author: Lisa Anne HendricksLisa Anne Hendricks is a credited author of Ethical and social risks of harm from Language Models.2021-12-08 · publication · Stated in source · Supported Inspect claim & evidence ↗Lisa Anne Hendricks: person page →
- Author: Mia GlaeseMia Glaese is a credited author of Ethical and social risks of harm from Language Models.2021-12-08 · publication · Stated in source · Supported Inspect claim & evidence ↗Mia Glaese: person page →
Documentary relations
Each relation is listed separately. Citing, crediting and reporting the use of a method are different acts, and none of them shows agreement or influence by itself.
Cited by · 5
Recorded documents that explicitly reference this one. A citation to a work does not identify which version was read.
- Training language models to follow instructions with human feedbackTraining language models to follow instructions with human feedback cites Ethical and social risks of harm from Language Models.2022-03-04 · publication · Stated in source · Supported Inspect claim & evidence ↗Training language models to follow instructions with human feedback: document page →
- Training Compute-Optimal Large Language ModelsTraining Compute-Optimal Large Language Models cites Ethical and social risks of harm from Language Models.2022-03-29 · publication · Stated in source · Supported Inspect claim & evidence ↗Training Compute-Optimal Large Language Models: document page →
- GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language ModelsGPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models cites Ethical and social risks of harm from Language Models.2023-03-17 · publication · Stated in source · Supported Inspect claim & evidence ↗Eloundou et al., GPTs are GPTs (2023): document page →
- Gemini: A Family of Highly Capable Multimodal ModelsGemini: A Family of Highly Capable Multimodal Models cites Ethical and social risks of harm from Language Models.2023-12-19 · publication · Stated in source · Supported Inspect claim & evidence ↗Gemini: A Family of Highly Capable Multimodal Models: document page →
- GPT-4 System CardGPT-4 System Card cites Ethical and social risks of harm from Language Models.Date unknown · Stated in source · Supported Inspect claim & evidence ↗GPT-4 System Card: document page →
Not recorded for this document: cites · explicitly credits · reports using a method from · method reported as used by · applies parts of · parts applied by · discusses · reports evaluation of. A missing record does not show that the relation does not exist.
What the document says it adds
The document’s own description of its contribution, quoted from the version each row names. A claim to be the first is the document’s claim; Selekoa does not assess it.
- In its own wordsThe authors state that the paper aims to help structure the landscape of risks associated with large-scale language models.“This paper aims to help structure the risk landscape associated with large-scale Language Models (LMs).” arXiv v1, PDF p. 1, abstract2021-12-08 · publication · Stated in source · Supported Inspect claim & evidence ↗
- In its own wordsThe paper outlines six risk areas: discrimination, exclusion and toxicity; information hazards; misinformation harms; malicious uses; human-computer interaction harms; and automation, access and environmental harms.“The paper outlines six specific risk areas: I. Discrimination, Exclusion and Toxicity, II. Information Hazards, III. Misinformation Harms, IV. Malicious Uses, V. Human-Computer Interaction Harms, VI. Automation, Access, and Environmental Harms.” arXiv v1, PDF p. 1, abstract2021-12-08 · publication · Stated in source · Supported Inspect claim & evidence ↗