My AI ethics reading list
These are the books, papers, and people who have shaped how I think about AI, and who think hardest about power, human dignity, and the people too often left out of the room.
It’s deliberately not a “neutral” list. I’ve foregrounded women of colour, Indigenous, queer, and disabled scholars, and voices from the Global South, because so often they saw the harms first and named them most clearly. It spans what the hype leaves out, too: AI’s environmental cost in energy, water, and minerals, and the hidden data labour it runs on. Read widely, argue with it, and pass on its suggestions. Book buy links support independent sellers — Readings (AU) and Bookshop.org (US).
Books
Algorithms of Oppression by Safiya Umoja Noble (2018)
How search engines encode racism and sexism, especially against Black women, under a veneer of neutrality.
Race After Technology by Ruha Benjamin (2019)
Names the “New Jim Code”: how tools that look objective quietly reproduce racial hierarchy.
Viral Justice by Ruha Benjamin (2022)
Part memoir, part manifesto: how small, everyday choices can compound into systemic change.
Unmasking AI by Joy Buolamwini (2023)
The founder of the Algorithmic Justice League on discovering the “coded gaze” and fighting it.
Artificial Unintelligence by Meredith Broussard (2018)
Coins “technochauvinism” — the belief that tech is always the answer — and dismantles it.
More Than a Glitch by Meredith Broussard (2023)
Bias in tech isn’t a glitch to be patched — it’s structural. A clear-eyed follow-up.
Automating Inequality by Virginia Eubanks (2018)
How automated decision systems surveil, profile, and punish poor and working-class people.
Weapons of Math Destruction by Cathy O’Neil (2016)
The original wake-up call on opaque, unaccountable algorithms operating at scale.
Atlas of AI by Kate Crawford (2021)
Maps the real costs of AI — the minerals, labour, energy, and data it extracts.
Data Feminism by Catherine D’Ignazio & Lauren F. Klein (2020)
A practical, intersectional framework for doing data work that confronts power.
Design Justice by Sasha Costanza-Chock (2020)
Designing with — not for — the communities most affected by what we build.
Dark Matters by Simone Browne (2015)
Roots today’s surveillance tech in the long history of monitoring Black life.
The Age of Surveillance Capitalism by Shoshana Zuboff (2019)
The definitive account of how our behaviour became raw material for prediction markets.
The Costs of Connection by Nick Couldry & Ulises A. Mejias (2019)
Introduces “data colonialism” — the extraction of human life as a new frontier.
Data Grab by Ulises A. Mejias & Nick Couldry (2024)
From the data-colonialism theorists: how Big Tech grabs our lives — and how to fight back.
Code Dependent by Madhumita Murgia (2024)
The FT’s AI editor tells the global, deeply human stories of living under AI systems.
Empire of AI by Karen Hao (2025)
Investigative reporting inside OpenAI and the new colonial logic of the AI race.
The AI Mirror by Shannon Vallor (2024)
A philosopher’s case that AI reflects our values back at us — and how to reclaim our humanity.
Papers & Reports
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?by Bender, Gebru, McMillan-Major & Shmitchell (2021)
The paper that named the risks of large language models — and cost two authors their Google jobs.
Gender Shades by Joy Buolamwini & Timnit Gebru (2018)
Showed commercial face analysis failed most on darker-skinned women. The audit that started a movement.
Datasheets for Datasets by Gebru, Morgenstern, Vecchione, Vaughan, Wallach, Daumé & Crawford (2018)
Proposes documenting where data comes from and its limits — now a transparency standard.
Model Cards for Model Reporting by Mitchell, Wu, Zaldivar, Barnes, Raji et al. (2019)
A simple template for disclosing how a model performs across different groups.
Actionable Auditing by Inioluwa Deborah Raji & Joy Buolamwini (2019)
Evidence that public audits actually push companies to fix biased systems.
Discriminating Systems: Gender, Race, and Power in AI by West, Whittaker & Crawford (2019)
Connects who builds AI to who it harms — the field’s diversity crisis is a bias crisis.
Excavating AI by Kate Crawford & Trevor Paglen (2019)
Pulls apart the training images that teach machines to classify people.
Anatomy of an AI Systemby Kate Crawford & Vladan Joler (2018)
A single map of one smart speaker’s full cost: minerals mined, energy burned, labour extracted.
The Environmental Impacts of AI — A Primer by Sasha Luccioni et al. · (2024)
Clear accounting of AI’s energy, water and mineral footprint across its whole lifecycle.
Making AI Less “Thirsty” by Li, Yang, Islam & Ren (2023)
Quantifies the freshwater that training and running large models quietly consumes.
The Misgendering Machines by Os Keyes (2018)
Shows how automatic gender recognition is built to erase trans and non-binary people.
From Pessimism to Promise by Payal Arora (2024)
What the Global South can teach the rest of us about designing inclusive technology.
Programmed Inequality by Mar Hicks (2017)
How Britain squandered its computing lead by pushing women out of the field.
Power and Progress by Daron Acemoglu & Simon Johnson (2023)
A thousand years of evidence on a single question: who actually benefits from new technology?
The New Age of Sexism by Laura Bates (2025)
How AI, deepfakes, sex robots, and the metaverse are reinventing misogyny for a new era.
Army of None by Paul Scharre (2018)
A defence insider on autonomous weapons and the choice to let machines decide who dies.
Cobalt Red by Siddharth Kara (2023)
The human and environmental toll of the cobalt mining in the DRC that powers our devices.
The Intersectional Environmentalist by Leah Thomas (2022)
You can’t protect the planet without uplifting its people — environmentalism meets justice.
Ghost Work by Mary L. Gray & Siddharth Suri (2019)
Exposes the hidden human workforce labelling and cleaning data behind “automated” systems.
Work Without the Worker by Phil Jones (2021)
How microwork outsources AI’s labour to the global poor, piece by invisible piece.
Behind the Screen by Sarah T. Roberts (2019)
The traumatic, hidden labour of the content moderators who keep platforms usable.
Queer Data by Kevin Guyan (2022)
Using — and resisting — gender, sex and sexuality data for queer liberation, not surveillance.
Rainbow Trap by Kevin Guyan (2025)
How the classifications meant to include LGBTQ people end up boxing them in.
Queer Data Studies edited by Patrick Keilty (2024)
Interdisciplinary essays rethinking how data extraction and modelling affect queer lives.
Against Technoableism by Ashley Shew (2023)
A disabled scholar dismantles the idea that technology should “fix” disabled people.
Disabling Intelligences by Rua M. Williams (2025)
Traces AI’s roots in eugenics and centres disabled experience to rethink intelligence itself.
Building Access by Aimi Hamraie (2017)
The politics of universal design — who gets to belong in the spaces and systems we build.
Indigenous Data Sovereignty and Policy edited by Walter, Kukutai, Carroll & Rodriguez-Lonebear (2020)
Global Indigenous scholars on reclaiming control of data — incl. Australian and Māori frameworks
Fairness for Unobserved Characteristics by Tomašev, McKee, Kay & Mohamed (2021)
Why fairness methods fail queer people: orientation and gender identity are often unmeasurable by design.
Queer in AI: A Case Study in Community-Led Participatory AI by Queer in AI (2023)
A model for building AI research led by the marginalised communities it affects.
What is the Point of Fairness? Disability, AI and The Complexity of Justice by Cynthia L. Bennett & Os Keyes (2019)
Argues disability shows why “fairness” isn’t enough — we need justice, not just equal treatment.
Disability, Bias, and AI by Whittaker, Alper, Bennett et al. (2019)
The landmark report on a missing conversation: how AI bias harms disabled people.
Algorithmic Injustice: A Relational Ethics Approach by Abeba Birhane (2021)
Argues ethics must centre the relationships and people affected, not abstract metrics.
The Values Encoded in Machine Learning Research by Birhane, Kalluri, Card, Agnew, Dotan & Bao (2022)
Analysed top ML papers to reveal the values — like performance over people — they quietly assume.
Making Kin with the Machines by Lewis, Arista, Pechawis & Kite (2018)
Indigenous epistemologies on relating to AI as kin, not as tools to dominate.
Out of the Black Box: Indigenous Protocols for AI by Abdilla, Kelleher, Shaw & Yunkaporta (2021)
Australian First Nations protocols and Country-centred design for building AI differently.
The CARE Principles for Indigenous Data Governance by Carroll et al. (2020)
Collective benefit, authority, responsibility, ethics — a standard for Indigenous data.
Indigenous Protocol and Artificial Intelligence edited by Lewis (2020)
Asks what AI built from Indigenous epistemologies and relationships could look like