I realize I say this about every AI book I read, but this one really is the best (so far), and most important in my view."Atlas of AI" by Kate Crawford is a well-researched work that should appeal to AI enthusiasts and opponents alike; not because it flatters either side, but because it challenges both to think beyond the usual narratives.

Whether you see Ai as a revolutionary tool for progress or a dystopian force of unchecked power, there’s no denying that it it is shaped by real-world systems of labor, industry, and politics.This book makes it clear: Ai is not just about algorithms and efficiency. It is about power: who wields it, who profits from it, and who is left to bear its costs.

For those who celebrate Ai’s potential (like me), Atlas of AI offers a sobering look at the material and ethical realities behind the inertia and hype.For those who critique AI as a damaging or dystopian force, the book provides a well-researched (eye-opening) foundation for those concerns.What makes it especially compelling is that it doesn’t fall into the trap of alarmism OR blind optimism.

Instead, Crawford takes us on a deep, methodical journey through the infrastructures that sustain artificial intelligence, revealing the hidden costs (labor, environmental, political) that come with EVERY so-called innovation.**Deconstructing the Myths of AI**One of the book’s greatest strengths is its ability to cut through the persistent myths surrounding artificial intelligence. Crawford systematically dismantles the notion that AI is a purely immaterial, frictionless technology.

She examines the vast mining operations necessary to produce hardware, the exploitative labor practices behind data annotation, and the enormous energy demands of AI training model. Ai, in her analysis, is not an autonomous or inevitable force—it is an industrial system deeply intertwined with capitalism, surveillance, and environmental degradation (much of her research applies to Big Tech, and not just Ai).

This perspective is crucial in an era where Ai is often presented as a revolutionary technology that exists outside of history and politics.Crawford makes it clear that Ai is not “just math” but a political tool wielded by those in power, often reinforcing existing inequalities.

The Ethics of Extraction and Control:

One of Crawford's most compelling arguments is the framing of AI as an extractive industry: one that harvests resources, labor, and data in much the same way as colonial enterprises have in the past.The book traces how Ai development is dependent on resource-intensive practices, from lithium mining for hardware to the invisible armies of low-wage workers tasked with cleaning and labeling data.

Crawford argues Ai is a system built on the extraction of value from the most vulnerable populations, whether they be gig workers, Amazon's "Mechanical Turk" laborers, or the communities living in the shadow of server farms that consume enormous amounts of water and energy.Crawfors cites numerous examples of how corporations like Google and Amazon, and even the government, skirt the system to save on taxes, while promising better futures to the resource-rich communities they exploit.

The theme of extraction extends beyond the physical to the digital realm.Crawford shows how personal data is commodified under the guise of “training AI,” reinforcing the asymmetrical relationship between those who generate data and those who profit from it.The book’s critique aligns with broader concerns about surveillance capitalism, demonstrating how Ai is often wielded as a means of control rather than liberation. (I learned some sad truths about local community policing and Ai)

AI and the Politics of Classification

Crawford explores how classification systems, often presented as objective/neutral, are deeply embedded with biases. Ai systems are trained on datasets shaped by human prejudices, yet are frequently deployed as infallible arbiters of truth. Crawford examines how facial recognition, predictive policing, and automated hiring systems encode and reinforce racial, gendered, and socioeconomic biases, often amplifying systemic discrimination.

This analysis is particularly relevant in today’s discussions on AI ethics. Crawford’s work underscores that Ai bias is not simply a technical glitch to be fixed, but rather a feature of the broader political and economic structures that Ai is designed to serve.

A Necessary and Timely Intervention

For those who have followed debates on Ai ethics, surveillance capitalism, and data justice, Atlas of AI provides a well-researched and compelling synthesis of these concerns, free from the noise we commonly hear on social media outlets. It is particularly valuable in challenging the mainstream, corporate-driven narratives that portray Ai as an inevitable and benign technological force.

Crawford’s writing is insightful, well-documented, and accessible, making complex ideas understandable without sacrificing depth. While the book is critical in tone, it does not merely scold Ai developers; rather, it offers a crucial intervention in ongoing discussions about how Ai is developed, deployed, and governed.The book had a surprisingly anti-capitalist/anti-technocratic tone, that inspired me to continue learning/aligning under the anti-fascist flag so many of us wield.

For artists, researchers, and technologists (especially those working at the intersection of Ai and creative expression) Atlas of Ai serves as a stark and vital reminder that technology is never neutral. It invites us to think critically about the systems we engage with and the ethical implications of our participation in Ai-driven ecosystems (and really, all major technologies).

Atlas of AI is an essential read for anyone interested in understanding the broader implications of artificial intelligence beyond the hype. It moves beyond discussions of algorithms and model accuracy to examine the power structures that shape Ai’s impact on society. By reframing AI as a material and political phenomenon rather than a disembodied technological marvel, Crawford provides a necessary course-correction to the dominant narratives surrounding Ai.

This book is not just for AI skeptics but for anyone who wants to engage in a deeper, more nuanced conversation about the technology shaping our present and future. If we are to meaningfully confront the challenges AI presents, we need more books that challenge us to think critically, demand accountability, and advocate for more just and equitable technological futures.

For those of us who engage with AI,whether as artists, researchers, developers, or critics, Atlas of AI should serve as a wake-up call. Too often, Ai artists defend the technology out of pride or personal investment, dismissing valid ethical concerns as fear-mongering. On the other side, anti-AI voices often resist engagement with nuance, preferring to frame Ai as an existential threat rather than a tool shaped by human systems of power. Both of these stances miss the point. Crawford makes it clear that the real battle isn’t Ai vs. artists or progress vs. tradition, it’s about who controls the technology, who benefits from it, and who is left to suffer the consequences. If we are serious about the future of art, technology, and creative autonomy, we must move beyond our egos and engage critically with the systems that shape Ai. This book gives us all a foundation to unify under, not in opposition to Ai itself, but in opposition to the unchecked power structures that exploit it and us.