Knowledge Centre Data & Society - News Knowledge Centre for legal, ethical and societal aspects of artificial intelligence and data-driven applications. en Sat, 04 Dec 2021 09:37:57 +0100 Sat, 04 Dec 2021 09:37:57 +0100 Tool: AI Blindspots healthcare Fri, 21 May 2021 08:00:00 +0200 (Willemien Laenens) How can you avoid replicating societal biases, prejudices and structural disparities in your AI healthcare system? In order to help you do this, the Knowledge Centre Data & Society developed the AI Blindspots healthcare card set. Tool: Digital Inclusion card game Fri, 21 May 2021 08:00:00 +0200 (Willemien Laenens) A role-playing game for companies and governments to get to know people's digital barriers and to test their own digital services. Tool: AI Blindspots 2.0 Sun, 06 Dec 2020 11:00:00 +0100 (Willemien Laenens) How can you take into account possible prejudices and structural inequalities before, during and after the development of your AI application? In order to help you do this, the Knowledge Centre Data & Society developed the AI Blindspots card set. Tool: AI Explainability 360 Open Source Toolkit Thu, 16 Apr 2020 13:31:00 +0200 (Cora van Leeuwen) A good way to get acquainted with the idea of explainable AI is to look at IBM Research's AI Explainability 360 Open Source Toolkit. In addition they offer multiple algorithms as a possible start of using explainable AI (XAI). Tool: Ethical OS Toolkit Tue, 31 Mar 2020 09:00:00 +0200 (Cora van Leeuwen) The Institute for the Future (IFTF) and the Omidyar Network have developed a toolkit to help predict difficult and unwelcome consequences and to prevent these from occurring while you develop products and projects based on AI. Data Ethics Canvas Tue, 31 Mar 2020 08:00:00 +0200 (Cora van Leeuwen) The Data Ethics Canvas provides tools to address ethical questions during the design and development of a project or product based on data, such as an AI application. Tool: The Tarot Cards of Tech Mon, 30 Mar 2020 16:00:00 +0200 (Cora van Leeuwen) A component of designing new technologies is imagining the impacy they will have on the world. In order to facilitate this Artefact created the Tarot Cards of Tech . These cards can. be used during team meeting to perform an informal impact assessment Aanpak begeleidingsethiek Mon, 30 Mar 2020 15:30:00 +0200 (Cora van Leeuwen) The approach to guidance ethics aims to start the dialogue about the connection between ethics and technology through a workshop. Artificiële Intelligentie Impact Assessment (AIIA)ële-intelligentie-impact-assessment-aiia Mon, 30 Mar 2020 14:23:00 +0200 (Cora van Leeuwen)ële-intelligentie-impact-assessment-aiia De Artificiële Intelligentie Impact Assessment (AIIA) is een gestructureerde methode om de (maatschappelijke) baten van een AI-systeem toepassing duidelijk in kaart te brengen. Daarnaast is er ook aandacht voor het analyseren van de betrouwbaarheid, veiligheid en transparantie van het AI-systeem. Tool: Aequitas Mon, 30 Mar 2020 14:01:00 +0200 (Cora van Leeuwen) The Aequitas tool performs an audit for your project. This method is intended to analyze whether there are prejudices in the data and in the models you use. You can perform the audit via the desktop or online tool. Tool: AI Maturity Tool Mon, 30 Mar 2020 13:00:00 +0200 (Cora van Leeuwen) The AI Maturity Tool is intended to find out to what extent your organization is ready to get started with AI. Tool: Building an Algorithm Tool Mon, 30 Mar 2020 11:00:00 +0200 (Cora van Leeuwen) With the Building an Algorithm Tool you ask ethical questions during the entire AI process: the design, development, test phase and implementation of the AI application. Tool: Data Ethics Guide Mon, 30 Mar 2020 10:00:00 +0200 (Cora van Leeuwen) The creators of the Data Ethics Guide want to highlight the issues surrounding ethics and make them manageable for companies that work with AI technology. They introduce different concepts of how ethics can be researched within a company. They do this through questions that help determine the current ethical situation. Tool: AI System Ethics Self-Assessment Tool Mon, 30 Mar 2020 09:00:00 +0200 (Cora van Leeuwen) Do you want to know how ethical your AI application is? The AI systems Ethics Self-Assessment Tool helps you to estimate four ethical principles yourself. Tool: People+ AI Guidebook Mon, 30 Mar 2020 08:00:00 +0200 (Cora van Leeuwen) The principles discussed in this guide are intended to assist user experience (UX) professionals and product managers with a human-centered approach to AI. This guide helps them put the user at the center of developing an AI application. Tool: Unbias Toolkit Sat, 28 Mar 2020 14:49:00 +0100 (Cora van Leeuwen) This toolkit created by and for young people is meant to share the online experience of young people with policy makers, regulators and the ICT industry. Tool: Principles for Accountable Algorithms en Social Impact Statement for Algorithms Fri, 27 Mar 2020 15:06:00 +0100 (Cora van Leeuwen) This tool by FAT/ML ( ‘Fairness, Accountability and Transparency in Machine Learning’) has as goal to promote the following ethical principles: responsibility, explainability, accuracy and accountability. Tool: Data Ethics Decision Aid (DEDA) Fri, 27 Mar 2020 14:50:00 +0100 (Cora van Leeuwen) This tool was created by and for data scientists. It consists of various methodologies that all can be applied at the start of a project. The aim is to document ethical issues and decisions in this regard, for example, to provide accountability towards stakeholders. Tool: Data Ethics Framework Fri, 27 Mar 2020 14:39:00 +0100 (Cora van Leeuwen) This ethical framework is intended for government agencies. However, it can also be useful for other organizations. Tool: SDoC for AI / AI service FactSheets Fri, 27 Mar 2020 14:36:00 +0100 (Cora van Leeuwen) The authors of this article state that suppliers of AI application must create a fact sheet for each product, demonstrating that the application is 'compliant' (an SDoC or 'supplier's declarations of conformity'), just like that for other products happens.