A working map

Reading.

Books, essays, papers, organizations, and concepts for thinking about AI anxiety.

Essays & Public Writing

Accessible writing that helps name the public, cultural, and emotional stakes of AI.

Ted Chiang

Clear writing on generative AI, compression, art, and meaning.

The Convivial Society

Technology, attention, limits, and human flourishing.

Sherry Turkle

Conversation, presence, and why face-to-face talk matters.

Ezra Klein

Public conversations that make AI legible without flattening the difficulty of the subject.

Helen Lewis

Cultural criticism that treats technology as a political and institutional event.

Henry Farrell & Cosma Shalizi

LLMs as cultural-technological objects rather than minds. Useful when the consciousness question is being asked too quickly.

Maciej Ceglowski

A sharp skeptical account of superintelligence discourse and one whole counter-position.

Jaron Lanier

A political-economy frame: AI as social collaboration mediated by software.

Erik Hoel

On what happens to meaning when generation becomes effectively free.

Research & Accountability

Technical and institutional work on model behavior, datasets, labor, and power.

Stochastic Parrots

Scale, data, bias, accountability, and fluent text.

Anthropic Interpretability

Research into what large language models are doing internally and how they can be inspected.

AI Now Institute

AI accountability, algorithmic management, surveillance, and power.

DAIR

AI research rooted in affected communities.

David Chalmers

A careful treatment of whether large language models could be conscious, and what would matter in the reasoning.

IIT 4.0

The current formulation of integrated information theory. Central to one corner of the machine-consciousness debate.

Risks from Learned Optimization

The mesa-optimization paper. Foundational for one strand of AI safety thinking.

Sparks of AGI

A useful artifact of how capability claims took shape: important for its rhetoric as well as its experiments.

Books

Longer works on risk, consciousness, systems, exhaustion, and relation.

Nick Bostrom

Superintelligence helped define the modern existential-risk debate. Useful, and best read critically.

Stuart Russell

A clear account of why optimizing fixed objectives can become dangerous when values are incomplete.

Brian Christian

A strong pre-LLM introduction to alignment as a human, technical, and institutional problem.

Kate Crawford

The material account: labor, minerals, water, electricity, geography, and infrastructure.

Hartmut Rosa

Resonance, acceleration, and the feeling of a world speeding up.

Karen Hao

Reporting on OpenAI as political economy, institutional history, and worldview.

Adam Becker

A science-journalism critique of longtermist and TESCREAL worldviews.

Byung-Chul Han

Exhaustion, self-optimization, performance culture, and the psychic costs of digital capitalism.

Iain McGilchrist

Attention, perception, reductionism, and the forms of intelligence a culture learns to reward.

David Chalmers

Digital life, simulation, consciousness, virtual worlds.

Donella Meadows

A practical way to think about models, incentives, infrastructure, and stories.

Talks & Lectures

Public explanations and position changes from people shaping the AI moment.

Hartmut Rosa

Lectures on resonance repay multiple watches, especially for thinking about acceleration and relation.

Iain McGilchrist

The short talks are an entry point; the longer lectures are where the argument lives.

Geoffrey Hinton

Important because someone central to the field publicly changed his stated position and explained why.

Murray Shanahan

Careful, philosophically literate AI research on consciousness and machines.

Timnit Gebru, Meredith Whittaker, Sara Hooker

Critical AI, policy, power, research culture, and political economy.

Collective Responses

Public statements, labor action, and organized responses to AI.

Pause Giant AI Experiments

A visible public intervention in the AI risk debate and an important historical marker.

Statement on AI Risk

A short statement that made existential risk publicly legible, while leaving other harms underdescribed.

Writers Guild of America

Labor protections around AI in a major creative industry.

Authors Guild

Consent, compensation, and the use of writers work in AI systems.

Concept Art Association

Artist advocacy around consent, training data, authorship, and creative labor.

SAG-AFTRA

A major labor struggle over synthetic performers, replicas, and consent.

Bletchley Declaration

State-level AI safety rhetoric is its own object of study.

Techno-Optimist Manifesto

A primary source for one worldview AIA is partly responding to.

TESCREAL

A critique of a bundled worldview: transhumanism, extropianism, singularitarianism, cosmism, rationalism, effective altruism, and longtermism.

Climate Psychology Alliance

A useful parallel: anxiety as a response to change at planetary scale.

Organizations & Projects

Institutions, projects, and venues that shape the broader ecosystem.

Mozilla Foundation

Civil-society work on internet health, trustworthy AI, privacy, and digital rights.

The Glass Room

Public education on data, surveillance, digital influence, and platform power.

Forensic Architecture

Technical tools used for investigation, public evidence, and accountability.

PauseAI

A movement advocating for slowing or pausing frontier AI development.

80,000 Hours

AI safety career resources, especially for readers coming from effective altruism communities.

Asterisk & Noema

Long-form writing on technology, civilization, governance, risk, and the future.

Center for AI Safety, FLI, ARC

The organized x-risk corner. Know what they actually argue, whether or not one agrees.

Eleos AI

A hub for the emerging field of AI welfare and possible moral patient status.

Long Now, Pioneer Works, e-flux

Adjacent cultural institutions working at the technology, culture, and future interface.

Concepts

Terms AIA returns to when discussing AI anxiety, agency, and human experience.

AI Anxiety

Emotional, existential, economic, and social unease produced by rapid advances in artificial intelligence.

Stochastic Parrot

Fluent text generated through statistical patterning without human-like understanding.

Alignment

The problem of building AI systems that reliably act within human values, intentions, and constraints.

Instrumental Convergence

The idea that different goals can lead advanced agents toward similar subgoals: resources, preservation, influence.

Goal Misgeneralization

A failure mode where a system appears to learn the intended behavior, then pursues the wrong objective in new conditions.

Mesa-Optimization

A situation where a trained system develops an internal objective that differs from the objective used to train it.

Mode Collapse

In generative systems, narrowed output diversity. More broadly, a useful metaphor for cultural homogenization.

Vocational Grief

Grief over changes to a profession, craft, or calling as skills and identities are devalued or automated.

Human-Only Conversation

Spaces where people speak with people about AI without AI mediation.

Integrated Information Theory

A theory of consciousness associated with Giulio Tononi, relevant to debates about artificial consciousness.

The Hard Problem

David Chalmers term for the difficulty of explaining subjective experience.

Predictive Processing

A family of theories that understand perception and action through prediction, error correction, and embodied inference.

TESCREAL

A contested term for a bundled worldview often present in AI futures discourse.

AI Welfare

The question of whether AI systems are or could become moral patients.

Constitutional AI, RLHF, RLAIF

Training methods that shape public arguments about model behavior, alignment, and accountability.

Anxiety as Information

AIA's stance that the felt response to the AI moment is not noise to be corrected, but signal to be read.

The Third Letter

AIA's deliberate ambiguity between anxiety, apocalypse, ambivalence and awareness, action, adaptation, acknowledgement.

Deep Cuts

Not first stops for every reader, but useful for expanding the intellectual range.

Ivan Illich

A foundational critique of tools and institutions that disempower the people they claim to serve.

Jacques Ellul

A major critique of technique: the drive to optimize everything according to technical efficiency.

Bernard Stiegler

Technology as something that shapes attention, memory, desire, and culture.

Yves Citton

Attention as an ecological and political resource.

Donna Haraway

Situated knowledge, responsibility, embodiment, and resistance to simple human-machine binaries.

Andy Clark

The extended mind, cognitive extension, and predictive processing.

Center for Humane Technology

Attention-economy critique, persuasive design, platform incentives, and social fragmentation.

Further Strands

Areas the library will keep developing as the project grows.

Lusophone and Portuguese voices

Portuguese-language AI policy, Iberian institutions, and older Portuguese thought on modernity and technology.

Religious and spiritual responses

The Vatican, Buddhist responses, Christian theological work, and broader spiritual interpretations of AI.

Mental-health practitioners

Therapists and clinicians writing about AI companionship, therapy bots, parasocial bonds, and care.

Indigenous and decolonial AI critique

Work on AI, land, sovereignty, data, language, extraction, and technical futures outside colonial defaults.

Children and AI

Childhood, school, play, attention, dependency, and the tools children are growing up with.

The considered counter-position

Arguments that alignment is solved, overstated, or the wrong frame, especially from inside the field.