➕​ More Resources

How can participants steer AI systems toward a common good? This workshop explored collective strategies for shaping outcomes in socio-technical systems from below. Keynote speakers, oral presentations and the list of accepted submissions are available on the workshop website, and on the website of NeurIPS. It is a great resource to get started within the field of ACA.

This resource was released while this tracker was in its final stages. It provides a curated list of well-documented examples on collective actions in algorithmic systems. It is a great, accessible resource to dive in some of the actions citizens can take. Our work on Algorithmic Collective Action completements the AI Resist List’s cases.

A tracker that focuses on how workers across the world mobilize in these fields that are highly impacted by the rise of Generative AI. The tracker’s design inspired ours.

A tracker of “traditional” action that gives a nice overview of what we call “Traditional Collective Action”. They propose a tracker that is open to contributions and a repository of resources for tech workers.

The AIAAIC Repository (standing for 'AI, Algorithmic and Automation Incidents and Controversies') is an independent, grassroots, public interest resource that collects and describes incidents and controversies driven by AI, algorithms and automation.

The Repository provides data and information on ethical misuses and the harms caused to users and other entities in order to increase the transparency and accountability of the perpetrators, and make the case for stronger and more proactive government policymaking.

The website also provides information on the methodology for data collection, and the taxonomies used to classify the different incidents, showing an interesting parallel with our project.

Algorithm registers are a type of tracker aiming to shed transparency on the use of algorithms to the general public. A very good resource to start with the topic is Soizic Pénicaud’s report, with almost 40 algorithmic registers mentioned. We mention here a couple other trackers, from a list started by Simon Chignard and expanded by our own references:

• Research articles on Algorithmic Registers

  Making Algorithm Registers Work for Meaningful Transparency. By S.Pénicaud with IA Ciudadana: A comprehensive report on algorithmic registers across governments.

  Understanding the Role of Algorithm Registers in AI Governance Through Comparative Analysis of China and the UK. (Yulu Pi, Wenlong Li, Jatinder Singh)

• Registers of government AI

  A directory of government AI registers maintained by the civic tech community.

• The ODAP (Observatoire des Algorithmes Publics)

  A tracker that focuses on the transparency of public sector algorithms in France and their respect of local regulations. It also proposes a blog section that engages in interviews and mediation work.

• Algorithm tips

  Algorithm register, similar to ODAP, in the US.

• Osservatorio dell’Amministrazione Automatizzata

  A registry of algorithms used by the Italian public administration.

• Tracking Automated Government Register UK

  A tracker of government algorithms in the United Kingdom.

• Tracking Automated Government Register Canada

  A tracker of government algorithms in Canada.

• AI Observatory

  A navigable database of automated Decision Making Systems in India, accompanied by reports and tools to help understand their ramifications.

• The Algorithm Register of the Dutch government

  The official register of algorithms used by the Dutch government.

• Algorithm Watch’s “Atlas de l’automatisation”

  A map of algorithms used in governments across Switzerland made by the Non-profit Algorithm Watch.

• Automating NYC

  An educational resource on Algorithms and Automated Decision Systems, that also provides a couple examples of such systems.

• Algorithmic Register of the City of Rennes, France

  This tracker is only one example of a "local" algorithmic register. This one in Rennes, France tracks more algorithms than the ODAP which focuses on algorithms used by the state at the national level, which shows how many algorithms there are in public services.

• The European Commission’s Public Sector Tech Watch

  A tracker that monitors the deployment of public AI services in the European Union (EU).

There are several legal questions raised by machine learning and, more recently, LLMs - for example regarding copyright and intellectual property. Some trackers can help to better follow the important amount of cases. We list a few here.

• DAIL – the Database of AI Litigation

  General purpose tracker for legal litigations around AI.

• The Artificial Intelligence and Copyright Case Tracker

  A tracker listing legal cases involving AI solutions and rights’ owners. Copyright is a highly topical issue with Generative AI systems.

A repository that documents “awful AIs” and contestation projects which inspired a few cases in the tracker. Citation: David Dao et al. (2022) “daviddao/awful-ai: Awful AI - 2021 Edition”. Zenodo. doi:10.5281/zenodo.5855972.

“Handbook for concerned citizens”, pointing them to many interesting sources and recommended actions. One of them is a list of anti-AI Tools that contains a few tools mentioned in our own tracker.

A repository of crowdsourced “collective datasets”, focusing on a multilingual effort that respects intellectual property. Users can share and license their datasets, and anyone can browse the repository.

A database focused on platform workers protests. A search feature allows to search by city, country and company name.

A database documenting police technology. It is a crowdsourcing initiative aiming to track and aggregate law enforcement tools of surveillance.

A hub of “resources for trade unions, labor organizations and worker-organizers fighting AI and automation at work”. As Alex Hanna puts it, the risk for workers is that "AI is not going to replace your job but it’s going to make your job much shittier". Among the resources are case studies that relate to topics mentioned in our taxonomy. Built by the Decentralized AI Research Institute (DAIR).

A website providing education material on common myths around Artificial Intelligence in an accessible way.

The AI Harm Map is a project built by the Ethical AI Alliance. It maps harms inflicted by AI systems across the world, mentioning landmark/textbook cases when it comes to the misuse of algorithmic systems. AI Harms are classified along 5 dimensions: Violation of human rights, Physical health/safety, Natural environment, Disinformation and Financial loss.