--- title: "Dario Amodei — author profile and papers" description: "Dario Amodei, Anthropic (United States): 2 papers on Elamur Research (2015–2016), h-index 32 — affiliations, topics, co-authors and who cites them in our…" canonical: "https://research.elamur.ai/free/en/authors/dario-amodei/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/dario-amodei/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Dario Amodei # Dario Amodei Dario Amodei's papers here span two areas of machine learning: [speech and spoken language](https://md.research.elamur.ai/free/en/topics/speech/), and the [bias, fairness and safety](https://md.research.elamur.ai/free/en/topics/bias-fairness-safety/) of AI systems. One line of work replaces hand-engineered recognition pipelines with a single neural network trained at scale. The other asks how systems built with today's learning methods can cause harm that no one intended. The most cited work is [Deep Speech 2](https://md.research.elamur.ai/free/en/papers/amodei-2015-deep-speech-2-end-end/) (2015). It trains one network of convolutional and recurrent layers with the CTC loss on thousands of hours of English and Mandarin audio. In doing so, it removes most of the separately tuned feature, acoustic, pronunciation and language components of a conventional recognizer. The paper also describes a training stack built on synchronous SGD across multiple GPUs, which brings full-scale training down to days. It sets "human-level" accuracy against transcriptions gathered from human workers. The later of the two papers, [Concrete Problems in AI Safety](https://md.research.elamur.ai/free/en/papers/amodei-2016-concrete-problems-ai-safety/) (2016), turns from building systems to their failure modes. It treats accidents in reinforcement and supervised learning as practical engineering problems rather than speculation about future agents. It sorts them by whether the objective is wrong, too expensive to check often, or breaks down during learning. From this it sets out five open problems, each illustrated with a hypothetical office-cleaning robot and paired with possible experiments. Co-authors on that paper include [Paul Christiano](https://md.research.elamur.ai/free/en/authors/paul-christiano/), [Jacob Steinhardt](https://md.research.elamur.ai/free/en/authors/jacob-steinhardt/) and John Schulman. Amodei is currently at Anthropic in San Francisco. Earlier bylines list Baidu, where the speech work was done with collaborators such as [Bryan Catanzaro](https://md.research.elamur.ai/free/en/authors/bryan-catanzaro/) and Adam Coates, as well as Stanford University and Princeton University. 2 papers on this site · 2015–2016 current affiliation [Anthropic (United States)](https://md.research.elamur.ai/free/en/companies/anthropic/), San Francisco Dario Amodei has 2 papers on this site, in 2015–2016, most of them on [Speech & spoken language](https://md.research.elamur.ai/free/en/topics/speech/), [Bias, fairness & safety](https://md.research.elamur.ai/free/en/topics/bias-fairness-safety/); often with [Adam Coates](https://md.research.elamur.ai/free/en/authors/adam-coates/), [Bryan Catanzaro](https://md.research.elamur.ai/free/en/authors/bryan-catanzaro/), [Carl Case](https://md.research.elamur.ai/free/en/authors/carl-case/). [OpenAlex](https://openalex.org/A5066197394) · [**h-index 32**](https://openalex.org/authors/A5066197394) · [53 works](https://openalex.org/works?filter=author.id:A5066197394) · [17 692 citations](https://openalex.org/works?filter=author.id:A5066197394&sort=cited_by_count:desc) OpenAlex, 2026-10-14 [2 works on this site](https://md.research.elamur.ai/free/en/authors/dario-amodei/#papers) · [2 689 citations in our library](https://md.research.elamur.ai/free/en/authors/dario-amodei/cited-by/) · 6 more works in our library The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Dario+Amodei) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Dario+Amodei) · [dblp](https://dblp.org/search?q=Dario+Amodei) · [ACL Anthology](https://aclanthology.org/search/?q=Dario+Amodei) ## Affiliations - [Anthropic (United States)](https://md.research.elamur.ai/free/en/companies/anthropic/) · San Francisco —current · 2021–2025 on 4 papers - [Machine Intelligence Research Institute](https://ror.org/01h4ass90) · Berkeley, United States —2022–2023 on 2 papers - [Stanford University](https://md.research.elamur.ai/free/en/universities/stanford-university/) · Stanford, United States —2006–2019 on 8 papers - [Baidu (China)](https://md.research.elamur.ai/free/en/companies/baidu/) · Beijing, China —2016–2018 on 2 papers - [Princeton University](https://md.research.elamur.ai/free/en/universities/princeton-university/) · Princeton, United States —2012–2017 on 5 papers From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 6 more institutions on single bylines are not listed. ## Topics [Speech & spoken language 1](https://md.research.elamur.ai/free/en/topics/speech/) · [Bias, fairness & safety 1](https://md.research.elamur.ai/free/en/topics/bias-fairness-safety/) ## Works by year ### 2010s [**2016** 1 paper](https://md.research.elamur.ai/free/en/papers/amodei-2016-concrete-problems-ai-safety/) · [**2015** 1 paper](https://md.research.elamur.ai/free/en/papers/amodei-2015-deep-speech-2-end-end/) ## Papers The most cited first (OpenAlex and our library). 1. **[Deep Speech 2: End-to-End Speech Recognition in English and Mandarin](https://md.research.elamur.ai/free/en/papers/amodei-2015-deep-speech-2-end-end/)** *(2015 · 2 302 citations)* Conventional speech recognizers are assembled from many hand-tuned parts, including feature extractors, acoustic, pronunciation and language models, and speaker adaptation. 2. **[Concrete Problems in AI Safety](https://md.research.elamur.ai/free/en/papers/amodei-2016-concrete-problems-ai-safety/)** *(2016 · 198 citations)* This 2016 position paper is about "accidents" in machine learning: harmful behaviour that nobody intended, caused by flawed design of real-world AI systems. ## Also in our library 6 more works of theirs are in our library without a page here yet; they will appear as we publish them. ## Co-authors here 10 co-authors with a page, the most papers together first. - [Adam Coates](https://md.research.elamur.ai/free/en/authors/adam-coates/) - [Bryan Catanzaro](https://md.research.elamur.ai/free/en/authors/bryan-catanzaro/) - [Carl Case](https://md.research.elamur.ai/free/en/authors/carl-case/) - [Erich Elsen](https://md.research.elamur.ai/free/en/authors/erich-elsen/) - [Greg Diamos](https://md.research.elamur.ai/free/en/authors/greg-diamos/) - [Jacob Steinhardt](https://md.research.elamur.ai/free/en/authors/jacob-steinhardt/) - [Jared Casper](https://md.research.elamur.ai/free/en/authors/jared-casper/) - [Jingdong Chen](https://md.research.elamur.ai/free/en/authors/jingdong-chen/) - [John Schulman](https://md.research.elamur.ai/free/en/authors/john-schulman/) - [Paul Christiano](https://md.research.elamur.ai/free/en/authors/paul-christiano/) ## Cited by Papers of our library cite their works 2 689 times (of the 140 884 papers whose bibliographies we read); [187 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/dario-amodei/cited-by/).