isHistory

Companion Publication · Living Catalog

Minds & Machines

Beyond the Series

Standalone essays extending the 78-article Minds & Machines series — the topics, profiles, and events the main series couldn't fully treat. From AI in healthcare and warfare to the lives of Judea Pearl and Rodney Brooks, from the 1943 McCulloch–Pitts paper to the January 2025 DeepSeek-R1 release.

🟡 Living Catalogby Ishaan90 pieces planned10 thematic clusters
90Pieces
0Published
10Clusters
3Types

Three flexible types

The companion publication publishes across three flexible types, with no fixed rotation. Each piece earns its own length and uses the format it needs.

Essays

B (essay) · 20 pieces

Thematic deep dives & current-affairs pieces on AI in society, political economy, technical sub-fields, and culture.

Profiles

B (profile) · 32 pieces

Biographical portraits of figures not profiled in the main series — the missing middle, RL, probabilistic AI, robotics, non-Western AI, critics, industry leaders.

Events

B (event) · 38 pieces

Standalone retrospectives on dated moments not covered in the main Events track — pre-Dartmouth, the 2013–15 deep-learning middle, robotics milestones, the post-2023 governance cascade.

Coming soon

The Beyond catalog is a living publication. The first standalone essays are being drafted now. Check back soon, or follow on Bluesky for publication announcements.

In the meantime, explore the main Minds & Machines series.

The complete catalog

All 90 pieces, organized by thematic cluster. Each piece is a standalone essay, profile, or event — designed to be read on its own, no prior context required.

Cluster 1 · AI in the Real World

Application-domain deep dives — where AI meets the professions, the industries, and the institutions.

7
  1. B01AI in Healthcare & Medicine: From MYCIN to AlphaFoldFrom MYCIN's 1974 non-deployment to AlphaFold to FDA-approved AI diagnostics — the 50-year story of AI in medicine and what comes next.essaylater
  2. B02AI in Warfare & the MilitaryProject Maven, Palantir, drone-targeting AI, and the autonomous-weapons treaty debate — how AI is reshaping modern warfare and military strategy.essaylater
  3. B03AI in Finance, Trading, and CreditAlgorithmic trading, Renaissance Technologies, high-frequency trading, and credit-scoring AI — how machine learning reshaped finance and fintech lending.essaylater
  4. B04AI in Legal Systems & Criminal JusticeCOMPAS, predictive policing, algorithmic sentencing — the story of AI in criminal justice and the ProPublica–Northpointe debate on algorithmic bias.essaylater
  5. B05AI in Education: Tutors, Detectors, and the Classroom BacklashKhanmigo, Duolingo, plagiarism detection, and the ChatGPT-in-the-classroom backlash — how AI is reshaping education and what learning is for.essaylater
  6. B06AI in Scientific DiscoveryAlphaFold, GNoME, AlphaGeometry, fusion-plasma control, and GraphCast — how AI became a research partner across biology, chemistry, math, and climate science.essaylater
  7. B07AI Assistants: The Siri/Alexa Decade (2011–2022)Siri, Alexa, Cortana, Google Assistant — the decade of consumer AI before ChatGPT, and how voice assistants bridged ELIZA to the LLM era.essaylater

Cluster 2 · The Political Economy of AI

Power, money, infrastructure — the material substrate of the AI revolution.

8
  1. B08The AI Chip Wars: Compute as the New OilTSMC, ASML, NVIDIA, US export controls, and the CHIPS Act — how the semiconductor supply chain became the geopolitics of AI.essaynext
  2. B09AI's Energy Bill: Data Centres, Water, and CarbonData-centre water use, training-run carbon footprints, the Microsoft–Three Mile Island restart — the environmental cost of the AI boom.essaynext
  3. B10The Invisible Workforce: AI Data Labour and Ghost WorkersMechanical Turk, Scale AI, Remotasks, Kenyan content moderators, RLHF raters — the invisible global workforce that trains every modern AI system.essaynext
  4. B11AI and Copyright: Who Owns the Training Data?NYT v. OpenAI, Getty v. Stability AI, Authors Guild suits — the copyright battles that will define who owns AI training data and at what cost.essaynext
  5. B12Open vs Closed: The Defining Fault Line of Modern AIHugging Face, LLaMA, Mistral, DeepSeek — the open-weights movement, the safety-vs-democratisation debate, and the future of AI governance.essaynext
  6. B13AI and Surveillance CapitalismShoshana Zuboff, Clearview AI, facial recognition, predictive policing — how surveillance capitalism shaped AI and what resistance looks like.essaylater
  7. B14AI and the Global SouthAI in Africa, Latin America, South Asia — indigenous-language NLP, the AI divide, and data extraction from the Global South.essaylater
  8. B15AI Nationalism and GeopoliticsThe US–China–EU triangulation, the UK and Japan national AI strategies, India's AI mission, and the rise of sovereign AI infrastructure.essaylater

Cluster 3 · The Missing Pioneers

Profiles that fill the 1965–1985 gap and the RL/probabilistic/robotics gaps.

9
  1. B16David Rumelhart: The Cognitive Face of ConnectionismDavid Rumelhart and the 1986 backpropagation paper — how the PDP Research Group at UC San Diego revived neural networks and founded modern connectionism.profilewhenever
  2. B17John Hopfield: The Physicist Who Bridged to Neural NetsJohn Hopfield's 1982 Hopfield network — how a Caltech physicist bridged statistical physics and neural computation, and won a Nobel Prize for it.profilewhenever
  3. B18Judea Pearl: Bayesian Networks, Causality, and the Turing AwardJudea Pearl's Bayesian networks, the do-calculus, and the 2011 Turing Award — the most consequential AI researcher invisible in the main series.profilelater
  4. B19Richard Sutton & Andrew Barto: The Fathers of Reinforcement LearningRichard Sutton and Andrew Barto — the fathers of reinforcement learning, the textbook that defined the field, and the third paradigm of AI.profilelater
  5. B20Rodney Brooks: Behavior-Based Robotics and the RoombaRodney Brooks — iRobot, the Roomba, MIT AI Lab, and behavior-based robotics. The most influential roboticist missing from the main series.profilewhenever
  6. B21McCulloch & Pitts: The 1943 Paper That Started EverythingWarren McCulloch and Walter Pitts — the 1943 paper that founded computational neuroscience, and the tragic life of Pitts, dead at 46.profilewhenever
  7. B22Donald Hebb: The Learning Rule That Built Modern AIDonald Hebb's 1949 Organization of Behavior — "fire together, wire together" — the conceptual ancestor of every modern learning rule.profilewhenever
  8. B23Paul Werbos: The 1974 Backpropagation ThesisPaul Werbos — the 1974 Harvard PhD that pre-dated Rumelhart–Hinton–Williams by 12 years, and the canonical credit-where-credit-is-due story.profilewhenever
  9. B24Kunihiko Fukushima: The Neocognitron and the Missing Japanese LinkKunihiko Fukushima's 1979 Neocognitron — the direct biological inspiration for LeCun's CNN, and the missing Japanese link in deep-learning genealogy.profilewhenever

Cluster 4 · The Deep Learning Middle, 2013–2015

Technical milestones between AlexNet (2012) and AlphaGo (2016).

10
  1. B25Word2Vec, August 2013: Vector Embeddings Become a Household ConceptWord2Vec — Mikolov's 2013 paper at Google that made vector embeddings a household concept and seeded the entire representation-learning programme.eventlater
  2. B26VAE, December 2013: The Foundational Generative PaperThe Variational Autoencoder — Kingma & Welling's December 2013 paper, one of the three foundational generative-model papers alongside GAN and diffusion.eventlater
  3. B27GAN, June 2014: The Bar-Argument InventionIan Goodfellow's GAN — invented in a Montreal bar in one evening, the foundational generative-model paper that powered five years of AI image research.eventlater
  4. B28Seq2seq, September 2014: Replacing Phrase-Based MT OvernightSequence-to-sequence learning — Sutskever et al.'s 2014 paper that replaced phrase-based machine translation overnight and prefigured the Transformer.eventlater
  5. B29Bahdanau Attention, September 2014: The Road to the TransformerBahdanau attention — the 2014 Montreal paper that introduced additive attention and quietly set up the 2017 Transformer breakthrough.eventlater
  6. B30DQN Atari, December 2013: The First General-Purpose Deep-RL SystemDeepMind's DQN Atari — the first general-purpose deep reinforcement learning system, and the technical ancestor of AlphaGo and the reasoning models.eventlater
  7. B31ResNet, December 2015: Skip-Connections and 100-Layer NetworksResNet — Kaiming He's residual-connection paper that won ILSVRC 2015, made 100-layer networks trainable, and is used by every modern deep network.eventlater
  8. B32Neural Style Transfer, August 2015: The AI-Art Movement BeginsNeural Style Transfer — Gatys et al.'s August 2015 paper that launched the AI-art movement and seeded the cultural conversation about generative creativity.eventlater
  9. B33TensorFlow Release, November 2015: Democratising Deep LearningTensorFlow — Google's November 2015 open-source release that democratised deep learning and became the production platform of choice.eventlater
  10. B34PyTorch Release, January 2016: The Dominant Research PlatformPyTorch — Meta's January 2016 release that became the dominant AI research platform and the foundation of the Hugging Face ecosystem.eventlater

Cluster 5 · AI Safety: The Discourse

The intellectual and institutional history of the AI safety movement.

8
  1. B35Nick Bostrom: Superintelligence and the Future of Humanity InstituteNick Bostrom's Superintelligence (2014) and the Future of Humanity Institute — how long-termist AI safety shaped OpenAI and Anthropic's founding.profilelater
  2. B36Eliezer Yudkowsky: MIRI, LessWrong, and "Shut It Down"Eliezer Yudkowsky — MIRI, LessWrong, the rationalist community, and the 2023 Time op-ed that called for shutting down all AI development.profilelater
  3. B37Paul Christiano: RLHF as AlignmentPaul Christiano — the Alignment Research Center founder who turned RLHF into the dominant alignment paradigm and shaped AI policy from inside.profilelater
  4. B38The CAIS "Statement on AI Risk," May 30, 2023The Center for AI Safety's May 2023 one-sentence statement — "mitigating extinction risk from AI should be a global priority" — signed by Hinton, Bengio, Altman.eventnext
  5. B39The OpenAI Board Crisis, November 17–22, 2023The OpenAI board crisis — five days that fired and reinstated Sam Altman, the most dramatic governance event in AI-industry history.eventnow
  6. B40The Bletchley Declaration and What Came After, November 2023The Bletchley AI Safety Summit — November 2023, the first heads-of-state-level AI summit, and the first US–China agreement on AI risk.eventnow
  7. B41The Future of Life Institute Pause Letter, March 22, 2023The FLI pause letter — March 2023, the open letter calling for a 6-month pause on giant AI experiments, signed by Elon Musk, Steve Wozniak, and 30,000 others.eventnext
  8. B42Anthropic Founding, 2021: The OpenAI SchismAnthropic's 2021 founding — the OpenAI schism that created the safety-focused frontier AI lab, and the "Buddhist charter" that shaped its culture.eventnext

Cluster 6 · Portraits of Dissent

Critics of AI across the decades — from Dreyfus to Bender.

10
  1. B43Hubert Dreyfus: What Computers Can't DoHubert Dreyfus — the Berkeley philosopher whose What Computers Can't Do (1972) became the most influential 20th-century critique of symbolic AI.profilewhenever
  2. B44John Searle: The Chinese RoomJohn Searle's 1980 Chinese Room thought experiment — the most-cited philosophical critique of strong AI and the question of whether syntax is semantics.profilewhenever
  3. B45Shoshana Zuboff: The Age of Surveillance CapitalismShoshana Zuboff's The Age of Surveillance Capitalism (2019) — the most-cited non-technical AI critic of the past decade on power, extraction, and resistance.profilelater
  4. B46Cathy O'Neil: Weapons of Math DestructionCathy O'Neil's Weapons of Math Destruction — the former Wall Street quant who exposed how algorithmic systems punish the poor and reinforce inequality.profilewhenever
  5. B47Joy Buolamwini: Gender Shades and the Algorithmic Justice LeagueJoy Buolamwini — the Gender Shades researcher who proved facial recognition fails on dark-skinned women, and founded the Algorithmic Justice League.profilelater
  6. B48Emily Bender & Margaret Mitchell: Stochastic ParrotsEmily Bender and Margaret Mitchell — the co-authors of Stochastic Parrots, the 2021 paper that named the LLM critique and reshaped AI ethics discourse.profilelater
  7. B49Meredith Whittaker: From AI Now to SignalMeredith Whittaker — AI Now co-founder, Google walkout organiser, and now president of the Signal Foundation, on surveillance, privacy, and AI power.profilewhenever
  8. B50Virginia Eubanks: Automating InequalityVirginia Eubanks's Automating Inequality (2018) — how algorithmic decision-making in welfare, child protection, and homelessness punishes the poor.profilewhenever
  9. B51Safiya Noble: Algorithms of OppressionSafiya Noble's Algorithms of Oppression (2018) — how search engine algorithms reinforce racism and sexism, and what search neutrality would require.profilewhenever
  10. B52Ruha Benjamin: Race After TechnologyRuha Benjamin's Race After Technology (2019) — the sociologist who framed AI bias as a feature, not a bug, of default technology design.profilewhenever

Cluster 7 · AI Beyond the West

Non-Anglospheric AI traditions — the largest single cultural gap.

7
  1. B53Victor Glushkov: Soviet Cybernetics and OGASVictor Glushkov — the Soviet cybernetician who designed OGAS, a networked economic planning system that prefigured the internet by a decade.profilewhenever
  2. B54Valentin Turchin: Refal and the Metasystem TransitionValentin Turchin — the Soviet dissident who invented Refal, founded metasystem-transition theory, and shaped émigré AI research after his exile.profilewhenever
  3. B55Kazuhiro Fuchi & Tohru Moto-oka: Japan's Fifth GenerationFuchi and Moto-oka — the leaders of Japan's $400M Fifth Generation Computer Systems project (1982–92), the largest state AI investment in history.profilewhenever
  4. B56Kai-Fu Lee: MSRA, Google China, AI SuperpowersKai-Fu Lee — Microsoft Research Asia founder, Google China president, and author of AI Superpowers, the central figure of Chinese AI's commercial rise.profilelater
  5. B57Liang Wenfeng: DeepSeek and the Open-Weight EarthquakeLiang Wenfeng — the hedge-fund founder behind DeepSeek, whose January 2025 R1 release triggered a US stock rout and reset the AI race.profilenext
  6. B58Arthur Mensch, Guillaume Lample & Timothée Lacroix: Mistral AIThe Mistral AI co-founders — three former Meta researchers who built Europe's frontier AI company on open weights and permissive licensing.profilelater
  7. B59Rediet Abebe: Black in AI, DAIR, and the Global SouthRediet Abebe — the Ethiopian-American researcher who co-founded Black in AI and DAIR, and shaped the global-South AI research agenda.profilewhenever

Cluster 8 · Robotics Across the Decades

The embodied-AI thread from 1950 to today.

10
  1. B60Embodied AI and Robotics: From Grey Walter to AtlasFrom Grey Walter's tortoises to Shakey to Boston Dynamics' Atlas — the 75-year story of embodied AI and why robotics is AI's hardest problem.essaylater
  2. B61Shakey the Robot, 1969–1972: The First Mobile RobotShakey the robot — SRI's 1969–72 mobile robot that perceived, planned, and acted, and how STRIPS became the foundation of AI planning.eventlater
  3. B62The DARPA Grand Challenges, 2004–2007The DARPA Grand Challenges — the 2004 Mojave race, Stanley's 2005 win, and the 2007 Urban Challenge that founded modern autonomous-vehicle research.eventlater
  4. B63Boston Dynamics BigDog / Atlas / Spot Reveals, 2005–2019Boston Dynamics' robot reveals — BigDog, Atlas, and Spot — the cultural milestones that shaped public expectations of robotics for two decades.eventlater
  5. B64Tesla Autopilot Launch, October 2014Tesla Autopilot — the October 2014 software update that put Level-2 self-driving in 60,000 Model S sedans and founded consumer autonomous driving.eventlater
  6. B65Waymo One Launch, December 2018Waymo One — the December 2018 launch of the first commercial robotaxi service in Phoenix, and the October 2020 driverless expansion.eventlater
  7. B66The Cruise Robotaxi Suspension, October 2023The Cruise suspension — October 2023, when California's DMV pulled Cruise's robotaxi permit after a pedestrian-dragging incident, and what it meant.eventlater
  8. B67Marc Raibert: Boston Dynamics and the Legged-Robot RevolutionMarc Raibert — the CMU and MIT roboticist who founded Boston Dynamics and built BigDog, Atlas, and Spot, the most visible robots of the era.profilewhenever
  9. B68Sebastian Thrun: Stanley, Google X, UdacitySebastian Thrun — the Stanford roboticist who won the 2005 DARPA Grand Challenge with Stanley, founded Google X, and launched Udacity.profilewhenever
  10. B69Cynthia Breazeal: Social Robotics and JiboCynthia Breazeal — the MIT roboticist who founded social robotics with Kismet and Jibo, and the case for robots as social companions rather than tools.profilewhenever

Cluster 9 · The LLM Era, 2018–2022

The pipeline from GPT-1 to ChatGPT — the technical runway of the revolution.

7
  1. B70GPT-1, June 2018: The Start of the GPT LineGPT-1 — OpenAI's June 2018 paper "Improving Language Understanding by Generative Pre-Training," the unsupervised pre-training debut that started the GPT line.eventnext
  2. B71BERT, October 2018: The Most-Cited NLP Model for Two YearsBERT — Google's October 2018 paper that set state-of-the-art on 11 NLP tasks at once and re-engineered Google Search around the Transformer.eventnext
  3. B72The GPT-2 "Too Dangerous to Release" Decision, February 2019GPT-2 — OpenAI's February 2019 staged-release decision, the first AI-safety theatre of the LLM era and the moment the staged-release playbook was born.eventnext
  4. B73Kaplan Scaling Laws, January 2020: "Just Add Compute"Kaplan scaling laws — OpenAI's January 2020 paper that gave the field its central prediction ("just add compute") and defined the 2020s AI race strategy.eventnext
  5. B74GPT-3, June 2020: The Pivot Point of the LLM EraGPT-3 — the June 2020 Brown et al. paper that pivoted the entire LLM era, introduced in-context learning, and made "foundation model" a category.eventnext
  6. B75InstructGPT, January 2022: RLHF Becomes Production AIInstructGPT — the January 2022 RLHF paper that made ChatGPT possible. Without it, ChatGPT would have been GPT-3 with a chat box.eventnext
  7. B76Chinchilla, March 2022: The Compute-Optimal ResetChinchilla — DeepMind's March 2022 compute-optimal scaling paper that reversed Kaplan's recipe (more data, smaller models) and reset training strategy.eventlater

Cluster 10 · Generative AI & Culture

How synthetic media reshaped art, law, and politics.

8
  1. B77The Colorado State Fair AI-Art Win, August 2022The Colorado State Fair — August 2022, when Jason Allen's "Théâtre D'opéra Spatial" became the first AI-generated image to win a fine-art competition.eventlater
  2. B78Stable Diffusion Release, August 2022Stable Diffusion — Stability AI's August 2022 open-source release that democratised image generation, triggered artist lawsuits, and seeded the open-AI debate.eventlater
  3. B79Sora Preview, February 2024Sora — OpenAI's February 2024 video-generation preview, the deliberate non-release that signaled AI video was the next multimodal frontier.eventlater
  4. B80Getty Images v. Stability AI, January 2023Getty v. Stability AI — the January 2023 lawsuit that will define the copyright regime for training image-AI models on copyrighted visual work.eventnow
  5. B81NYT v. OpenAI & Microsoft, December 2023NYT v. OpenAI — the December 2023 lawsuit that will likely define whether training LLMs on copyrighted articles is fair use or infringement.eventnow
  6. B82Authors Guild v. Google, October 2015Authors Guild v. Google — the October 2015 Second Circuit ruling that established fair use for book scanning and now underpins every LLM training defence.eventlater
  7. B83Synthetic Media, Deepfakes, and the Crisis of EvidenceDeepfakes, voice cloning, and the "liar's dividend" — how synthetic media collapsed photographic evidence and launched a detection-versus-generation arms race.essaylater
  8. B84The AI Election, RevisitedThe 2024 AI election — the New Hampshire Biden robocall, deepfakes, and the liar's dividend. What happened, what didn't, and lessons for 2026–28.essaynow

Additional high-priority pieces

Pieces that emerged from the gap analysis as urgent but don't fit cleanly into the 10 thematic clusters.

6
  1. B85DeepSeek-R1 and the Reset of the AI Race, January 2025DeepSeek-R1 — the January 2025 release that matched OpenAI's o1 at $6M training cost, triggered a $589B NVIDIA stock rout, and reset the AI compute debate.eventnow
  2. B86The EU AI Act RolloutThe EU AI Act — March 2024 passage, August 2024 entry into force, and the 2025–26 enforcement milestones that will define Europe's AI regulatory regime.eventnow
  3. B87Hallucinations: Why AI Makes Things UpAI hallucinations — why large language models fabricate facts, what causes it, what RAG and attribution can fix, and what the industry still can't solve.essaynext
  4. B88Prompt Engineering and the New Art of Talking to MachinesPrompt engineering — chain-of-thought, few-shot, ReAct, structured prompts, and the rise (and partial fall) of the "prompt engineer" as a profession.essaynext
  5. B89Altman Senate Testimony, May 16, 2023Sam Altman's May 2023 Senate testimony — the first time a major AI CEO endorsed federal licensing for AI, and the moment that reset the regulatory conversation.eventnext
  6. B90The White House Voluntary Commitments, July 21, 2023The White House voluntary commitments — July 2023, when seven AI companies pledged safety standards at the Rose Garden, the first industry-wide AI safety pact.eventnext

Editorial calendar

Pieces sequenced by topicality and time-sensitivity, not by importance.

Now

7

Time-sensitive (within 1 month)

Next

19

Topically hot (1–3 months)

Later

43

Evergreen, slow-burn (3–12 months)

Whenever

21

Evergreen, no deadline

How Beyond differs from the main series

DimensionMain SeriesBeyond the Series
Structure5 acts, 26 rotations, 3 tracksNone — each piece is standalone
Length~8,000 words per articleVariable (1,500–12,000 words)
ToneNarrative, story-driven historyVariable — essay, reportage, biography
CadenceWeekly rotationOn publication, no fixed schedule
Reader contractClosed at 78 articlesOpen-ended, living catalog
URL pattern/articles/minds-and-machines/[slug]//blog/beyond/minds-and-machines/[slug]/

Editorial principles

  1. Each piece stands alone. No piece assumes the reader has read any other piece. Cross-links are invitations, not prerequisites.
  2. Each piece earns its own length. Some stories need 1,500 words; some need 12,000. The 8,000-word main-series format is not the default.
  3. Each piece uses the format it needs. Tick-tock timelines for fast-moving events. Long-read narrative for intellectual biographies. Reported essays for ongoing stories. Retrospectives for settled history.
  4. Each piece links back to the main series. The main series is the canon. Beyond pieces extend it; they do not replace it.
  5. No piece is required reading. The main series is the complete story. Beyond is for readers who want more.