{"work":{"id":"b4ce1c45-ef69-445a-a872-dbb785b485e9","openalex_id":null,"doi":null,"arxiv_id":"2112.04359","raw_key":null,"title":"Ethical and social risks of harm from Language Models","authors":null,"authors_text":"Laura Weidinger, John Mellor, Maribeth Rauh, Conor Griffin, Jonathan Uesato, Po-Sen Huang","year":2021,"venue":"cs.CL","abstract":"This paper aims to help structure the risk landscape associated with large-scale Language Models (LMs). In order to foster advances in responsible innovation, an in-depth understanding of the potential risks posed by these models is needed. A wide range of established and anticipated risks are analysed in detail, drawing on multidisciplinary expertise and literature from computer science, linguistics, and social sciences.\n  We outline six specific risk areas: I. Discrimination, Exclusion and Toxicity, II. Information Hazards, III. Misinformation Harms, V. Malicious Uses, V. Human-Computer Interaction Harms, VI. Automation, Access, and Environmental Harms. The first area concerns the perpetuation of stereotypes, unfair discrimination, exclusionary norms, toxic language, and lower performance by social group for LMs. The second focuses on risks from private data leaks or LMs correctly inferring sensitive information. The third addresses risks arising from poor, false or misleading information including in sensitive domains, and knock-on risks such as the erosion of trust in shared information. The fourth considers risks from actors who try to use LMs to cause harm. The fifth focuses on risks specific to LLMs used to underpin conversational agents that interact with human users, including unsafe use, manipulation or deception. The sixth discusses the risk of environmental harm, job automation, and other challenges that may have a disparate effect on different social groups or communities.\n  In total, we review 21 risks in-depth. We discuss the points of origin of different risks and point to potential mitigation approaches. Lastly, we discuss organisational responsibilities in implementing mitigations, and the role of collaboration and participation. We highlight directions for further research, particularly on expanding the toolkit for assessing and evaluating the outlined risks in LMs.","external_url":"https://arxiv.org/abs/2112.04359","cited_by_count":null,"metadata_source":"pith","metadata_fetched_at":"2026-06-29T07:43:13.530068+00:00","pith_arxiv_id":"2112.04359","created_at":"2026-05-09T05:50:28.431041+00:00","updated_at":"2026-06-29T07:43:13.530068+00:00","title_quality_ok":true,"display_title":"Ethical and social risks of harm from Language Models","render_title":"Ethical and social risks of harm from Language Models"},"hub":{"state":{"work_id":"b4ce1c45-ef69-445a-a872-dbb785b485e9","tier":"hub","tier_reason":"10+ Pith inbound or 1,000+ external citations","pith_inbound_count":81,"external_cited_by_count":null,"distinct_field_count":13,"first_pith_cited_at":"2022-01-20T15:44:37+00:00","last_pith_cited_at":"2026-06-23T16:46:36+00:00","author_build_status":"not_needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"not_needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-06-29T14:38:56.672063+00:00","tier_text":"hub"},"tier":"hub","role_counts":[{"context_role":"background","n":27},{"context_role":"other","n":2}],"polarity_counts":[{"context_polarity":"background","n":24},{"context_polarity":"unclear","n":3},{"context_polarity":"support","n":2}],"runs":{"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-14T17:49:48.848454+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"On the Opportunities and Risks of Foundation Models","work_id":"a18039e9-928d-47c9-a836-32656a71bf71","shared_citers":16},{"title":"Scaling Language Models: Methods, Analysis & Insights from Training Gopher","work_id":"47ce8be9-e500-407d-af41-ac2d132215eb","shared_citers":12},{"title":"LaMDA: Language Models for Dialog Applications","work_id":"1b66d0a5-f6ae-4332-8025-c662dc64b238","shared_citers":11},{"title":"Red Teaming Language Models with Language Models","work_id":"d1274c54-508f-42f9-aeb3-91db13f3a622","shared_citers":11},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":10},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":10},{"title":"Training Compute-Optimal Large Language Models","work_id":"b2faf28d-86b7-429c-bc42-469458efc246","shared_citers":9},{"title":"Training language models to follow instructions with human feedback","work_id":"52aff42f-4fa9-4fcf-bdb3-1459b9bebf65","shared_citers":9},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":9},{"title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","work_id":"ed240a10-5b19-406c-baa5-30803f465785","shared_citers":8},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":8},{"title":"Universal and Transferable Adversarial Attacks on Aligned Language Models","work_id":"3322fa86-1768-4677-8425-dd326b45e078","shared_citers":8},{"title":"Chain-of-Thought Prompting Elicits Reasoning in Large Language Models","work_id":"d1cf6693-a082-403c-ada9-dac7b96341f9","shared_citers":7},{"title":"Improving language models by retrieving from trillions of tokens.Preprint arXiv:2112.04426","work_id":"c8e7ce21-2f18-4c10-b0cf-16cd3574fe33","shared_citers":7},{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","shared_citers":7},{"title":"PaLM: Scaling Language Modeling with Pathways","work_id":"a94f3ef7-2c49-4445-93fe-6ec16aafd966","shared_citers":7},{"title":"Quantifying Memorization Across Neural Language Models","work_id":"35487ec1-b90b-4ace-95bd-1bce30064b2e","shared_citers":7},{"title":"Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned","work_id":"1aabd84d-3779-4ba9-ba2f-15ce264a9b1e","shared_citers":7},{"title":"WebGPT: Browser-assisted question-answering with human feedback","work_id":"e25ef3e1-4848-4cb9-bf28-67a420591165","shared_citers":7},{"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","shared_citers":6},{"title":"GLU Variants Improve Transformer","work_id":"17d0763c-1016-41ab-a478-478e890765eb","shared_citers":6},{"title":"M., and Bowman, S","work_id":"6f78b350-8a2a-4d00-9090-41971c52baaf","shared_citers":6},{"title":"S entence P iece: A simple and language independent subword tokenizer and detokenizer for neural text processing","work_id":"81a6320b-c2e1-4d74-a03e-9e1ff6bbed8d","shared_citers":6},{"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","shared_citers":6}],"time_series":[{"n":10,"year":2022},{"n":2,"year":2023},{"n":2,"year":2024},{"n":1,"year":2025},{"n":27,"year":2026}],"dependency_candidates":[]},"error":null,"updated_at":"2026-05-14T17:49:55.323192+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"items":[{"title":"Qwen3 Technical Report","outcome":"unchanged","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"counts":{"fixed":0,"merged":0,"unchanged":1,"quarantined":0,"needs_external_resolution":0},"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-14T17:49:55.255272+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Ethical and social risks of harm from Language Models","claims":[{"claim_text":"This paper aims to help structure the risk landscape associated with large-scale Language Models (LMs). In order to foster advances in responsible innovation, an in-depth understanding of the potential risks posed by these models is needed. A wide range of established and anticipated risks are analysed in detail, drawing on multidisciplinary expertise and literature from computer science, linguistics, and social sciences.\n  We outline six specific risk areas: I. Discrimination, Exclusion and Toxicity, II. Information Hazards, III. Misinformation Harms, V. Malicious Uses, V. Human-Computer Inte","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Ethical and social risks of harm from Language Models because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T17:49:55.330323+00:00"}},"summary":{"title":"Ethical and social risks of harm from Language Models","claims":[{"claim_text":"This paper aims to help structure the risk landscape associated with large-scale Language Models (LMs). In order to foster advances in responsible innovation, an in-depth understanding of the potential risks posed by these models is needed. A wide range of established and anticipated risks are analysed in detail, drawing on multidisciplinary expertise and literature from computer science, linguistics, and social sciences.\n  We outline six specific risk areas: I. Discrimination, Exclusion and Toxicity, II. Information Hazards, III. Misinformation Harms, V. Malicious Uses, V. Human-Computer Inte","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Ethical and social risks of harm from Language Models because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"On the Opportunities and Risks of Foundation Models","work_id":"a18039e9-928d-47c9-a836-32656a71bf71","shared_citers":16},{"title":"Scaling Language Models: Methods, Analysis & Insights from Training Gopher","work_id":"47ce8be9-e500-407d-af41-ac2d132215eb","shared_citers":12},{"title":"LaMDA: Language Models for Dialog Applications","work_id":"1b66d0a5-f6ae-4332-8025-c662dc64b238","shared_citers":11},{"title":"Red Teaming Language Models with Language Models","work_id":"d1274c54-508f-42f9-aeb3-91db13f3a622","shared_citers":11},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":10},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":10},{"title":"Training Compute-Optimal Large Language Models","work_id":"b2faf28d-86b7-429c-bc42-469458efc246","shared_citers":9},{"title":"Training language models to follow instructions with human feedback","work_id":"52aff42f-4fa9-4fcf-bdb3-1459b9bebf65","shared_citers":9},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":9},{"title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","work_id":"ed240a10-5b19-406c-baa5-30803f465785","shared_citers":8},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":8},{"title":"Universal and Transferable Adversarial Attacks on Aligned Language Models","work_id":"3322fa86-1768-4677-8425-dd326b45e078","shared_citers":8},{"title":"Chain-of-Thought Prompting Elicits Reasoning in Large Language Models","work_id":"d1cf6693-a082-403c-ada9-dac7b96341f9","shared_citers":7},{"title":"Improving language models by retrieving from trillions of tokens.Preprint arXiv:2112.04426","work_id":"c8e7ce21-2f18-4c10-b0cf-16cd3574fe33","shared_citers":7},{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","shared_citers":7},{"title":"PaLM: Scaling Language Modeling with Pathways","work_id":"a94f3ef7-2c49-4445-93fe-6ec16aafd966","shared_citers":7},{"title":"Quantifying Memorization Across Neural Language Models","work_id":"35487ec1-b90b-4ace-95bd-1bce30064b2e","shared_citers":7},{"title":"Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned","work_id":"1aabd84d-3779-4ba9-ba2f-15ce264a9b1e","shared_citers":7},{"title":"WebGPT: Browser-assisted question-answering with human feedback","work_id":"e25ef3e1-4848-4cb9-bf28-67a420591165","shared_citers":7},{"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","shared_citers":6},{"title":"GLU Variants Improve Transformer","work_id":"17d0763c-1016-41ab-a478-478e890765eb","shared_citers":6},{"title":"M., and Bowman, S","work_id":"6f78b350-8a2a-4d00-9090-41971c52baaf","shared_citers":6},{"title":"S entence P iece: A simple and language independent subword tokenizer and detokenizer for neural text processing","work_id":"81a6320b-c2e1-4d74-a03e-9e1ff6bbed8d","shared_citers":6},{"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","shared_citers":6}],"time_series":[{"n":10,"year":2022},{"n":2,"year":2023},{"n":2,"year":2024},{"n":1,"year":2025},{"n":27,"year":2026}],"dependency_candidates":[]},"authors":[]}}