Biblioteca
-
Science in the age of large language models
26 de Abril de 2023Abeba BirhaneAI fairness, accountability, and transparencyFour AI ethics and policy experts highlight the potential risks associated with LLMs along with the need for responsible usage and careful consideration.
-
AI Transparency in Practice
20 de Março de 2023Mozilla Insights, Ramak Molavi Vasse'i, Jesse McCroskyAI Transparency in Practice Builders on what works — and what doesn't. Mozilla’s research on AI transparency (lead by Ramak Molavi Vasse'i), with practical advice from Thoughtworks.
-
Creating community-driven datasets: Insights from Mozilla Common Voice activities in East Africa
7 de Março de 2023Jan KrewerAI fairness, accountability, and transparency / Internet health / Community building / Voice technologyThe report summarizes lessons-learnt and strategies of three voice communities used to create publicly available datasets. It is based on interviews with people driving these efforts, including Mozilla Fellows, community coordinators, contributors as well as supporting donor organizations.
-
Is that even legal? A guide for builders experimenting with data governance
15 de Fevereiro de 2023Maximilian Gahntz, Mozilla InsightsData governance & stewardshipThe data economy is not built on a level playing field — but it can be fixed!
-
Beyond Single-Mindedness: A Figure-Ground Reversal for the Cognitive Sciences
10 de Janeiro de 2023Abeba BirhaneAI fairness, accountability, and transparencyA letter to the editors of the Cognitive Science Journal from a transdisciplinary collective.
-
ChatGPT, Galactica, and the Progress Trap
9 de Dezembro de 2022Deb Raji, Abeba BirhaneAI fairness, accountability, and transparencyFor WIRED, the authors hone in on the need for accountability in LLM deployment.
-
Towards Responsible Recommending
7 de Dezembro de 2022Maximilian GahntzAI fairness, accountability, and transparency / Platform accountabilityRecommendations to platforms and policymakers on how to move towards a more responsible recommending ecosystem.
-
Handling and Presenting Harmful Text in NLP Research
1 de Dezembro de 2022Abeba BirhaneAI fairness, accountability, and transparencyThe article provides advice for NLP practitioners, with concrete steps for mitigating harm in research and in publication, on how to handle, present, and discuss harmful texts.
-
Digital Futures in Mind
1 de Novembro de 2022Piers GoodingInternet healthThis research was conducted as part of Gooding's 2020 Mozilla fellowship.
-
Digital Care and Philanthropy: Findings and Recommendations
27 de Outubro de 2022AmarelaPrivacy, security & trackingThis report was developed by researchers Foz and Amarela with the support of FASE – Federação de Órgãos para Assistência Social e Educacional and the Mozilla Foundation's Tech + Society Fellowship.
