Manifesto on "Feminism, Pacifism and Artificial Intelligence"

Why? How was developed? Basic principles Detailed principles Who is it aimed at? Persons who have signed

Why this Manifesto?

All international organisations, led by the United Nations, recognise that despite the advances in gender equality that have taken place in recent years, there is still much work to be done and many steps to be taken by society as a whole. This is why we need to take on board feminist perspectives in order to criticise gender inequality and other forms of social injustice. On the other hand, international organisations themselves are systematically calling for an immediate global ceasefire and an end to wars and genocide in all corners of the world in order to focus on the real struggle, which is climate change. Hence we need pacifism for the critique of war and other large-scale violence. Feminism and pacifism therefore share common ground in their struggle against oppression and violence, both structural, cultural and direct, and in their promotion of a more just and equitable society. Feminist perspectives are known to have broadened the understanding of pacifism by showing how violence and war are closely linked to patriarchal power structures, while pacifism has enriched feminism by offering a vision of nonviolent resistance as a path to social transformation.

After a series of social changes and gains during the second half of the 20th century, we are perplexed by a 21st century reality where advances related to artificial intelligence (AI) not only reflect a gender bias present in our society, but do so in a much more amplified and discriminatory way. To begin with, AI technology is often equated with the power of men, while other genders are presented as less technologically skilled and less interested than their male counterparts. Given that this situation reinforces the structural hierarchies and stereotypes of society, which is based on pre-assigned gender roles, we are in danger of perpetuating an axis of oppression identified and contested in recent decades. It is therefore necessary to adopt an intersectional approach that addresses issues of gender, age, ethnicity, socio-economic status and other determinants, in addition to adopting a human rights-based approach to AI governance that takes into account transparency, accountability and human dignity.

At the same time, the collusion between the arms industry and the big tech sector reveals how the use of powerful AI systems in wars and conflicts not only raises a series of moral and legal questions about the relationship between people and machines, but also reveals a highly destructive capacity that reminds us of the worst episodes of the 20th century. In this context, it should be remembered that more than 30 years since the end of the Cold War, the US and Russia keep many of their intercontinental ballistic missiles on alert and, far from dismantling them, are equipping themselves with the help of AI. All this is happening despite the incorporation of feminist perspectives into peacekeeping operations and UN security discourses. But it is clear that there is also an effort in the opposite direction that, among others, uses technology as a catalyst for its positions. The clearest and most current example is in Palestine where, through the Lavender AI system, thousands of Palestinians – mostly women and children or people who were not involved in the fighting – are being killed by Israeli air strikes due to a use of AI that only accelerates the destruction of humanity.

In this sense, individuals, organisations and initiatives that prioritise a feminist and pacifist perspective in the use of AI from day one have a greater capacity for transformative innovation to reduce and not perpetuate violent conflict and, of course, to address, albeit indirectly, the structural relationship between sexism, violence and warmongering. Clearly, this also entails eliminating the gender gap on its own or in combination with other dimensions such as those centred on ethnicity, culture and age, as well as socio-economic class, disability or sexuality. The centrality of these issues can not only favour a set of AI potentials, but also a less apocalyptic and much more positive vision of the future of humanity, which can only be realised through feminist and pacifist perspectives. This requires interdisciplinary solidarity and dialogue between diverse fields to inspire hope and action, to foster a shared vision that becomes more socially acceptable and contributes to the achievement of common goals. Thus, feminism and pacifism, in an AI context, in their joint commitment to a more just and nonviolent society, are not only mutually reinforcing, but also offer a powerful framework for challenging the forms of violence and oppression that persist in today’s world.

This manifesto underlines the need, first, for AI to be designed and developed with feminist perspectives in mind, and thus for a range of actions to be taken to implement equitable representation along the dimensions of gender, age, ethnicity and diverse abilities, with a special focus on silenced voices. Secondly, peace must be promoted, violence avoided and environmental sustainability contributed to, with or without AI. Third, it is essential to detect, correct or mitigate biases in AI systems to avoid discrimination and promote equality and social justice in all technological applications. Fourth, it is necessary to foster a culture of transparency around AI, with external audits and ethics committees, respecting human rights and ensuring the security of users’ personal data. Finally, it is essential to educate on social justice, human rights and technological ethics, and to promote global justice in a more determined way. The social reality surrounding AI is perhaps the most clearly global phenomenon today, with long and unequal supply chains, where the role of transnational corporations, the geopolitical spin-off in the race for AI and the offshoring of much of the environmental, eco-social and social impacts is increasingly visible. This means that collaboration between countries, small businesses, academic institutions and non-profit organisations is needed to ensure a fair distribution of the benefits of AI.

How was this Manifesto developed?

The elaboration has followed three sequential phases: proposal, discussion and revision. The participatory process of constructing the manifesto began with a proposal phase where relevant previous experiences were identified and consolidated through the promoting organisations. In this stage, initial ideas were contributed on the contents to be included in the manifesto and different people were proposed, taking into account their expertise, knowledge and backgrounds, in order to validate the initial ideas and obtain new proposals to enrich the document.

After sharing all this information about the rationale and main themes of the manifesto, a shared proposal was made through dialogue, debate and joint reflection that was organised in various groups using the World Café methodology. This is designed to create a safe and welcoming environment in which to intentionally connect multiple ideas and perspectives on an issue by engaging multiple people in several rounds of small group conversations. It is a simple, effective and flexible format for sharing knowledge, addressing the complexities and nuances of an issue or issues, and creating possibilities for action in groups of various sizes, based on the notion of group intelligence. In this second phase of discussion, the chosen topics were shared and favoured the building of interdisciplinary alliances between feminist and pacifist advocates in the ethical development of AI, as well as a common agenda that prioritises non-violence, justice and equality in the development and deployment of AI technologies. Therefore, social science contributions were included as an essential element in order to identify and implement scientific evidence of social impact, those that improve the lives of citizens and, at the same time, those that protect the most vulnerable people. This discussion phase in which the different perspectives obtained from the focus groups were evaluated, debated and compared was key to reaching the first list of points in the manifesto, highlighting those points that required more consensus or deeper analysis. Of course, the discussion also served to identify areas of disagreement or where ideas needed to be refined, ensuring that each contribution was critically reviewed. Finally, following the discussions, a third phase of participatory review was opened, where all stakeholders have the opportunity to provide final comments and make adjustments to the content of the manifesto. This stage focuses on ensuring that the resulting document is inclusive and representative of the diverse voices involved, and consistent with the initial objectives of the manifesto itself. The aim of this final part of the manifesto process is to end with the collective approval of the manifesto’s contents, establishing a solid and participatory basis for its future development.

We therefore encourage all stakeholders, from public administration to industry, academia and civil society, to support our manifesto on Feminism, Pacifism and Artificial Intelligence by signing up to it.

Basic principles of the Manifesto (in 5 points)

Ensure that intersectional feminist perspectives shape the design, development and use of AI, promoting justice, equity and the fight against all forms of oppression and discrimination.
Promote pacifism in the development and use of AI by prohibiting the design, sale and use of AI systems intended for autonomous lethal weapons or any technology that contributes to armed violence, ensuring that AI is developed with a firm commitment to peace and nonviolent conflict resolution.
Establish mechanisms for transparency and accountability in the use and development of AI, ensuring audits and access to public information on data, methodologies, and algorithms used, in order to avoid bias and promote democratic governance of AI.
Ensure that AI projects respect human rights, peace and sustainability, avoid contributions to conflict or violence, and prioritise the use of renewable energy in AI infrastructure.
Create initiatives to increase the representation of marginalised groups in AI development by providing resources, training and support, while incorporating ethics, social justice and global justice into AI literacy.

The detailed principles of the Manifesto (in 50 points)

Ensure that intersectional feminist perspectives guide AI design and functionality decisions, promoting justice and equity.
Use data and models that fairly represent all people and intersections, avoiding simplification and stereotyping.
Implement guidelines to ensure that AI applications promote gender equality and combat sexism, with special emphasis on child protection, as well as all other axes of oppression and discrimination.
Foster the inclusion of racialised people of different backgrounds and abilities, strengthening collaboration and responsible innovation in AI design and development processes and governance.
Systematically use methods of detection and correction of bias to avoid discrimination of individuals through ex-ante and ex-post audits and controls.
Ensure that all areas of AI development, from research to commercialisation, have more pluralistic and equitable gender, age, ethnic and cultural representation.
Create safe and accessible spaces for under-represented communities to express and influence their views on the design, development and governance of AI.
Develop initiatives to provide resources, training and targeted financial support for entrepreneurship for groups historically discriminated against in the field of AI technology.
Ensure that each AI project is developed with an explicit commitment to non-aggression or negative impact on peace and human rights, avoiding contributing to conflict or violence in any form.
Include courses and workshops that teach AI developers the principles of pacifism, nonviolence, human rights, non-discrimination, nonviolent conflict resolution and the importance of considering them at all times.
Establish a clear prohibition on the development, sale and use of AI systems applied to weaponry that can act autonomously, in order to avoid contributing to armed violence and violent conflict management.
Harness AI’s analytical capacity to identify and warn of potential situations of violence or humanitarian emergencies, taking into account the full enjoyment of human rights.
Organise international forums and conferences that bring together people from different backgrounds to discuss how AI technology can be used to promote peace rather than the other way around, in order to prevent global violence.
Encourage and support the creation of feminist and peace coalitions for AI development and governance that promotes equality and non-violent conflict resolution.
Establish registration mechanisms in order to make public the methodologies, data sources and algorithmic processes used in all AI development.
Implement systems that record every action and decision taken by AI systems to enable full external audits.
Establish independent ethics committees to review and approve all new AI systems to ensure that they are established with high ethical and human rights standards.
Ensure that users have control over their data and that they can revoke consent themselves at any time, including the necessary tools to enable parents or guardians to do so in the case of minors.
Implement state-of-the-art security technologies to protect users’ data from unauthorised access or use.
Demand measures for feminist cybersecurity to fight against sexist and misogynist violence on the networks, for privacy of information or equal access to the use, but also to the development, of technology.
Ensure that all AI projects comply with applicable laws and the highest ethical standards of data protection at local and international level.
Inform users when they interact with AI-generated content and explain the implications of this interaction.
Create a mandatory civil liability code of conduct that all AI professionals must follow.
Implement validation processes to ensure that AI products comply with all legal and technical requirements before being made available on the market.
Ensure commitment to human rights and AI standards in order to protect and promote the rights and freedoms of individuals.
Analyse how AI can perpetuate or mitigate existing inequalities and include this analysis as part of the impact study to advance a fairer use of technology.
Incorporate case studies and real examples to illustrate how AI can affect differently depending on the socio-economic and cultural context.
Articulate mechanisms to ensure the active participation of affected communities so that they can express their needs and concerns, and can directly influence the development of the IA.
Implement systems that monitor and report on the social impact of AI with a comprehensive approach, especially in vulnerable communities, and taking into account the invisible consequences.
Establish laws to ensure compensation or remedial measures for communities that may be negatively impacted by IA projects.
Offer fellowships and mentoring programmes designed to increase the representation of under-represented groups in technology fields.
Call for measures to respect the labour rights of people working in the training of these technologies, usually women in countries of the global South.
Create regular forums where AI practitioners can receive direct feedback from the public and other stakeholders.
Establish programmes that not only teach technical skills with AI, but also promote an understanding and promotion of social justice and global justice.
Develop interdisciplinary curricula that equip students with the tools to address complex ethical and human rights issues within technology.
Emphasise the need to maintain human control over critical decisions, especially in areas such as criminal justice or resource management.
Analyse cases where AI has served to both promote gender equality and jeopardise it, and strategies for aligning with these movements.
Promote a critical view of how technology is designed and implemented, fostering a deeper understanding of the material consequences with an ecofeminist vision.
Establish a moratorium on data centres until they implement energy efficiency strategies and operate with renewable energy without energy monopolisation.
Prohibit the location of data centres in areas and territories where there is a clear and evident drought and water scarcity that affects the very communities that live there and their natural habitats.
Conduct environmental impact assessments on a regular basis, introducing independent stakeholder compliance guarantees for any new AI technology or upgrade.
Create clear guidelines to ensure that data used in AI respects copyright and intellectual property.
Require that programmes, actions, regulations or any type of initiative linked to AI incorporate scientific evidence with social impact in order to make scientifically informed decisions that contribute to gender equality and the eradication of violence in the use of AI.
Promote practices that minimise the carbon footprint and optimise the use of resources in the production and maintenance of AI technologies.
Require companies to disclose the environmental impact of their AI operations, energy consumption, water consumption and carbon emissions, so that users are aware of the environmental impact of their use.
Promote initiatives that use AI to improve access to resources and opportunities for excluded communities, without creating new digital exclusions.
Allocate resources to AI projects that offer practical solutions to everyday problems, such as optimising energy efficiency.
Promote international collaborations that allow the flow of knowledge and resources between countries and cultures, avoiding the concentration of technological knowledge.
Establish policies that encourage the participation of small and medium-sized enterprises, startups, academic institutions and non-profit organisations in AI research, monitoring and development.
Strengthen organisations working in the Catalan sphere on algorithmic justice and cybersecurity from feminist coordinates, raise awareness among specific audiences (journalists, educators, etc.) and promote the generation of feminist data with consistent information systems.

Who is it aimed at?

Government and public administration departments.
International organisations that set global standards and policies.
Non-governmental organisations (NGOs).
Trade unions and employers.
Technology companies.
Smaller innovative start-ups.
Universities and research centres.
Industry associations in general and tech associations in particular.
Associations of professional bodies.
Media.
Foundations and other bodies.
Community and grassroots organisations.
Companies from different economic sectors
Scientific associations.

Persons who have signed

Adrià Menéndez i Molis. PhD student, UPC – Centre for Research in Economics and Agri-Food Development
Adrià Molina Rodríguez. Pre-doctoral researcher in Computer Vision
Alba Crespo López. Pre-doctoral researcher, University of Barcelona
Albert Sabater. Observatory of Ethics in Artificial Intelligence of Catalonia
Alexandra Lillo. Observatory of Ethics in Artificial Intelligence of Catalonia
Alfons Sabater Coll. Coordination Area Records Mútua Colaboradora Social Security
Alicia de Manuel. Expert analyst in ethical AI
Alma D. Calderón López. Sociologist and researcher
Ana Belén Rojo Ojados. ESDAPC, professor of plastic arts and design, department of humanities
Ana Burgués-Freitas. University of Granada
Anaïs Varo. University of Girona
Anna Arias Duart. Researcher
Anna Escobedo. University of Barcelona
Anna Pérez Rodríguez. Engineer working in the third sector
Anna Sàez Bernal. Political scientist specialising in public management
Anna Sala. Design
Anna Sardà. Teno Education
Antonio Madrid. University of Barcelona
Aquilino Díaz Méndez. Retired teacher at IES Alfonso II, Oviedo
Arlet Brufau i Centelles. Observatory of Ethics in Artificial Intelligence of Catalonia
Asturias AEBE
Atia Cortés Martínez. Researcher
Aura Hernández Sabaté. Autonomous University of Barcelona
Bahar Kılıç Adilçe. Assistant professor in the Department of Film and Television
Beatriz Villarejo. Autonomous University of Barcelona
Benjamín Menéndez. Teacher
Benjamin Menéndez Martínez. Teacher
Blanca Febré López. Secondary School Teacher
Carla Illas Jorba
Carles Planas. El Periódico
Carles Serrat Piè. Polytechnic University of Catalonia-BarcelonaTECH
Carlos Bajo. Oxfam
Carmen José Villarreal Redón. Social Educator
Carmen Ripoll Ribera. Teacher of Vocational Training Courses
Carol Romero. Decidim.org
Carolina Grau del Valle. University of Valencia
Claudia Hereu Rico. Research
Cristina Pulido. Autonomous University of Barcelona
Cristina Sanchez. University of Girona
Cynthia. Engineer
Dario Castañé. Rizoma
David Blanco de Tena-Dávila. Lecturer – researcher. International University of Catalonia
EA Lluvia D’amore González Gallizio. Pedagogue, Vocational Training Lecturer
EduAlter
Elisenda Gozalo Salellas. Lecturer. Open Institute of Catalonia
Enric Senabre Hidalgo. University of Barcelona
Camí del Cros School
Esther Roca Campos. University lecturer and researcher
Francisco Orgilés Pérez. Primary school teacher
Frederic Guerrero-Solé Pompeu Fabra University, +RAIN Film Festival
Gemma Pedret Fontova. UPC
Gina Tost Faus. Technology and innovation specialist
Giovanna Lara Lecturer – researcher
GRBIOIO. Research Group in Biostatistics and Bioinformatics
Guadalupe Gómez Melis. Polytechnic University of Catalonia. Professor of Statistics and Operational Research and member of the Ethics Committee
Guillermo Legorburo Torres. International University of Valencia
Guiomar Rovira. University of Girona
Héctor García Morales. Physicist and science communicator
Héctor Martínez Romero. Educational advisor
IRIS AEBE Educational Association Based on Scientific Evidence
Jaume Feliu. Teacher
Joan Manuel del Pozo Álvarez. Professor Emeritus at the University of Girona
Joan Solé Pla. University of Girona PDI
Joana Sabater Castelló. Biotechnology student
Jordi Ascensi Sala. Andorra Research and Innovation, member of the Bureau of the AI Committee of the Council of Europe
Jordi Manuel Antolí Martínez. University professor and researcher. University of Alicante
Jordi Moner Roma. Teacher
Jordi Ocaña Rebull. Retired university professor
Jose Manuel Bobadilla Gómez. Sociologist of science and technology
Josep Àngel Guimerà. Autonomous University of Barcelona
Josep Maria Canal. VIU / University of Valencia
Judith Membrives i Llorens. Researcher
Karlos Castilla. Catalan Institute of Human Rights
Karma Peiró. Freelance journalist
La Consultoria Social
La familia digital
Lafede.cat – Organisations for Global Justice
Laia Abelló. Communications technician
Laia Orozco Joan. Head of Digital Equity Laboratory, Fundació Jaume Bofill
Laia Riera Armengol. Family doctor
Laia Serra. Barcelona Bar Association
Lena de Botton. University of Barcelona
Lídia Arroyo. Lecturer at the Autonomous University of Barcelona
Lidia Bordanoba. PhD student
Lídia Puigvert. University of Barcelona
Liliana Arroyo. Specialist in digital innovation and social impact
Lola S. Almendros. Postdoctoral researcher in STS Studies
Lucía Gallego. University of the Basque Country UPV/EHU
Luisa Herreras Blanco. Secondary school teacher
Magdalena Camps Pardo. Teacher
Manel Soengas i Núñez. Education. ICT Area
Mar Joanpere Foraster. Rovira i Virgili University
Marc Carulla. University of Barcelona
Marc González Pedret. Dietitian-Nutritionist (Association of Dietitians-Nutritionists of Catalonia)
Margarida Romero. International University of Catalonia
María Carmen Rufete López. Early childhood educator UPV
María Eugenia. Researcher
Maria Lacueva Lorenz. Open University of Catalonia
Maria Marfil Gutiérrez. Sociology student
María Paz Anderson Madariaga. FACEPA
Marta Galceran. CIDOB
Marta Martinez Canal. Digital content strategist
Marta Soler Gallart. University of Barcelona
Marttina Breittmayer. Student
Mercè Gisbert Cervera. Rovira i Virgili University
Miguel Rebollo Pedruelo. Polytechnic University of Valencia
Miquel Rubio. Computer scientist
Montse Bosch. Polytechnic University of Catalonia. Ethics Committee Montserrat Clua Autonomous University of Barcelona
Nadia Zougagh. Student of Law and Political Science and Administration
Nagla Rizk. Professor of Economics & Director, Access to Knowledge for Development, Onsi Sawiris School of Business, The American University in Cairo
Nia Soler. Illustrator and writer
Núria Espuny Salvadó. Engineer
Núria Marín García. Lecturer Olga Serradell Autonomous University of Barcelona
Olivia Blanchard. Consultant and researcher
Oriol Ríos González. Rovira i Virgili University
Oxfam Intermón
Patrícia Ventura. Independent researcher and consultant
Paula Petrone. Barcelona Institute for Global Health
Paulina Sicius.
Pedro Cárcamo Petridis. Doctor (c), University of Barcelona
Pere Pèries Escòlies. Composer and playwright
Pilar Morera Basuldo. University of Girona
Rafael López Vidal. Bookseller
Ramon López de Mántaras Badia. Emeritus Professor of Research at the CSIC and member of the Institute of Catalan Studies
Raquel Aguilar Mora. Teacher
Reis Gallego Perales. Vocational training teacher and master’s student in e-learning
Roger Campdepadrós Cullell. University of Girona
Rommy Morales – Olivares. University of Barcelona
Rosa Cuscó Martinez. Communication
Rosa Garvín Fernández. Retired teacher
Rubén Rodríguez Llopis. Teacher
Sandra González Campos. Optometrist
Sandra Montserrat Buendia. University of Alicante
Sara Carbonell Sevilla. Professor, University of Valencia
Sergi Luque Manjón. Teacher
Silvia Rueda Pascual. Computer Engineer
Sílvia Simon Rabasseda. IQCC, University of Girona
Catalan Statistical Society (SoCE)
Susana Gómez Bretones. Primary School Teacher
Round Table of Third Sector Social Entities of Catalonia
Teresa López-Pellisa. Laboratory for Future Studies. University of Alcalá
Teresa Sordé Martí. Autonomous University of Barcelona
Teresa Vázquez Peris. Teacher
Thais Ruiz de Alda. Technologist. Founder of Digitalfems
Toni Bertólez Martínez. Pre-doctoral researcher in Physics
Ulises Cortés. Polytechnic University of Catalonia. Professor of Artificial Intelligence
Vanessa Monar del Castillo. Teacher
Vicenç Casas Sagarra. Engineer and teacher
Víctor Manuel Marí Sáez. University of Cádiz
White Rose Research Centre. Social impact of education, feminism, masculinities and overcoming gender-based violence
Xavier Cardona Iguacen. Bioethicist
Xiana Bueno. Researcher, School of Public Health, Indiana University

Manifesto on “Feminism, Pacifism and Artificial Intelligence”

Join the Manifesto