Article
Form, meaning and language models
Bender and Koller distinguish linguistic form from meaning in debates about models' language understanding.
Why read it?
Helps make questions about what it means for AI to “understand” language more precise.
Keep in mind
The article defends an argument about language technology. Later systems require their own assessment.
Reference details
Title as recorded: Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data
Author(s): Emily M. Bender; Alexander Koller · Year: 2020
Publication / host: Association for Computational Linguistics
Source language: English
DOI: 10.18653/v1/2020.acl-main.463
KP-R-BBD00C6F76F8
Legislation
The Belgian Constitution
The official Dutch-language constitutional text, including Article 149 on giving reasons for judgments.
Why read it?
A primary legal source for questions about explanation and accountability in judicial decisions.
Keep in mind
How a provision should be interpreted and applied requires a separate assessment.
Reference details
Title as recorded: De Belgische Grondwet
Author(s): Not recorded · Year: Not recorded
Publication / host: www.senate.be
Source language: Dutch
KP-R-BFDEFD7DA27F
Article
Black boxes in black-hole imaging
A philosophical study of methods whose workings are difficult to understand, used to produce images of black holes.
Why read it?
Helps examine how much understanding of a method is needed to trust its output.
Keep in mind
The analysis concerns one scientific practice. It does not offer a general judgement on AI.
Reference details
Title as recorded: Black boxes in black hole imaging
Author(s): Juliusz Doboszewski; Jamee Elder · Year: 2026
Publication / host: Springer Science and Business Media LLC
Source language: English
DOI: 10.1007/s11229-026-05751-2
KP-R-C03BCDB8A542
Article
How our expectations of AI change
A philosophical approach to how expectations of responsible behaviour change through interactions between people and AI.
Why read it?
Explores how people and AI systems influence one another's use and assessment.
Keep in mind
The article develops ideas. It does not prescribe a universally applicable code of conduct.
Reference details
Title as recorded: A coevolutionary account of normative human–AI interaction
Author(s): Jörg Noller · Year: 2026
Publication / host: Springer Science and Business Media LLC
Source language: English
DOI: 10.1007/s44163-026-02130-1
KP-R-C1CC35C3C78D
Article
Simpson: when a comparison reverses
Blyth examines how a comparison within groups can reverse when those groups are combined.
Why read it?
Provides further explanation of the tension between subgroup figures and an aggregate picture.
Keep in mind
A statistical pattern does not automatically identify a cause. Additional assumptions are needed.
Reference details
Title as recorded: On Simpson's Paradox and the Sure-Thing Principle
Author(s): Colin R. Blyth · Year: 1972
Publication / host: Informa UK Limited
Source language: English
DOI: 10.1080/01621459.1972.10482387
KP-R-C2B6B62614ED
Article
Science with hard-to-understand models
Duede asks what scientists can understand when the workings of a deep-learning model are difficult to inspect.
Why read it?
Explores when a useful model also provides insight.
Keep in mind
The article makes a philosophical assessment. Predicting well and understanding well remain different questions.
Reference details
Title as recorded: Deep Learning Opacity in Scientific Discovery
Author(s): Eamon Duede · Year: 2023
Publication / host: Cambridge University Press (CUP)
Source language: English
DOI: 10.1017/psa.2023.8
KP-R-C2B89A1F493B
ArticleCounterpoint
Is dependence on AI really new?
A philosophical article challenges the claim that relying on AI in science creates an entirely new problem for knowledge.
Why read it?
Offers a counterpoint to the idea that dependence for knowledge began with AI.
Keep in mind
Read this argument alongside research identifying possible new risks.
Reference details
Title as recorded: Of opaque oracles: epistemic dependence on AI in science poses no novel problems for social epistemology
Author(s): Jakob Ortmann · Year: 2025
Publication / host: Springer Science and Business Media LLC
Source language: English
DOI: 10.1007/s11229-025-04930-x
KP-R-C3927C92725A
Classic text
Vajirā: a whole and its parts
A Buddhist text compares the name given to a whole with the parts that make it up.
Why read it?
Offers a historical starting point for asking whether a whole is more than its parts.
Keep in mind
The original comparison concerns a chariot. A bicycle version is a modern adaptation.
Reference details
Title as recorded: Vajirā en de vergelijking met een wagen
Author(s): Not recorded · Year: Not recorded
Publication / host: www.dhammatalks.org
Source language: English
KP-R-C47D86B2A138
Perspective
Hanson: where are the extraterrestrial civilisations?
Hanson explains the Great Filter as a possible account of the absence of visible extraterrestrial civilisations.
Why read it?
Helps distinguish a possible explanation from an established finding.
Keep in mind
This is speculative reasoning that relies heavily on assumptions. It does not predict humanity's future with certainty.
Reference details
Title as recorded: Hanson: The Great Filter
Author(s): Not recorded · Year: Not recorded
Publication / host: hanson.gmu.edu
Source language: English
KP-R-C597EDD16460
Chapter
Newcomb: choosing after someone has predicted your choice
Nozick discusses a choice in which reasoning from a prediction may recommend something different from reasoning about your influence on the outcome.
Why read it?
An original source for Newcomb's conflict between two approaches to rational choice.
Keep in mind
How the predictor works and how reliable it is form part of the problem's assumptions.
Reference details
Title as recorded: Newcomb’s Problem and Two Principles of Choice
Author(s): Robert Nozick · Year: 1969
Publication / host: Springer Netherlands
Source language: English
DOI: 10.1007/978-94-017-1466-2_7
KP-R-C845BD3A5E0D
Article
Hardin: when everyone uses a shared resource
Hardin describes the tension between individual benefit and preserving a resource that people share.
Why read it?
An original source for asking how individual advantage can lead to collective loss.
Keep in mind
Read Ostrom too. Unmanaged open access differs from use governed collectively by a community.
Reference details
Title as recorded: The Tragedy of the Commons
Author(s): Garrett Hardin · Year: 1968
Publication / host: American Association for the Advancement of Science (AAAS)
Source language: English
DOI: 10.1126/science.162.3859.1243
KP-R-CAF754510DD4
Article
Re-examining hindsight
This study revisits the explanation that new knowledge changes our memory of earlier judgements.
Why read it?
Shows how an explanation of hindsight can be investigated again.
Keep in mind
The researchers test a particular explanation with an adapted design. Other explanations remain possible.
Reference details
Title as recorded: Hindsight Bias Through Knowledge Updating: A Conceptual Replication of Groß et al. (2023)
Author(s): Barbara K. Kreis; Antje Hermann; Thorsten Pachur; Julia Groß · Year: 2025
Publication / host: University of California Press
Source language: English
DOI: 10.1525/collabra.147499
KP-R-CBD582600629
Reference work
The prisoner's dilemma explained
An overview of the prisoner's dilemma: how individual interests can conflict with the benefits of cooperation.
Why read it?
Introduces a familiar problem about self-interest and cooperation.
Keep in mind
The model has explicit assumptions. It does not explain every form of human cooperation.
Reference details
Title as recorded: Het prisoners’ dilemma uitgelegd
Author(s): Not recorded · Year: Not recorded
Publication / host: plato.stanford.edu
Source language: English
KP-R-CD1E277964E0
Article
ELIZA: a program that appears to converse
Weizenbaum describes how his simple ELIZA program can create the impression of a conversation.
Why read it?
Helps ask why we may experience understanding in a conversation with a computer.
Keep in mind
ELIZA is historical software. Modern language models work differently.
Reference details
Title as recorded: ELIZA—a computer program for the study of natural language communication between man and machine
Author(s): Joseph Weizenbaum · Year: 1966
Publication / host: Association for Computing Machinery (ACM)
Source language: English
DOI: 10.1145/365153.365168
KP-R-CDDB0C54767C
ArticleStarting point
To trust or to think?
A study examines interventions that prompt people to think before following AI advice.
Why read it?
Provides research on how to create room for independent judgement.
Keep in mind
The results come from particular decision tasks. Their relevance to other working environments needs to be checked.
Reference details
Title as recorded: To Trust or to Think
Author(s): Zana Buçinca; Maja Barbara Malaya; Krzysztof Z. Gajos · Year: 2021
Publication / host: Association for Computing Machinery (ACM)
Source language: English
DOI: 10.1145/3449287
KP-R-CDF0CB7F1BAA
Guidance
Microsoft on intended AI behaviour
An example of the principles an AI provider uses to describe how a system should behave.
Why read it?
Helps compare stated principles with observable behaviour.
Keep in mind
Microsoft is describing its own policy. This is not an independent guarantee that systems follow it.
Reference details
Title as recorded: Een gedragscode voor AI van Microsoft
Author(s): Not recorded · Year: Not recorded
Publication / host: microsoft.ai
Source language: English
KP-R-CFD9572E957A
Article
Making automated decisions reviewable
An approach to making decisions, processes and responsibilities open to later investigation and challenge.
Why read it?
Provides background for asking what information is needed to contest a decision.
Keep in mind
The approach describes how review can be enabled. Practical arrangements still need to be put in place.
Reference details
Title as recorded: Reviewable Automated Decision-Making
Author(s): Jennifer Cobbe; Michelle Seng Ah Lee; Jatinder Singh · Year: 2021
Publication / host: ACM
Source language: English
DOI: 10.1145/3442188.3445921
KP-R-D031D3B9077A
Chapter
Robots and human responsibility
In a provocatively titled chapter, Bryson argues that responsibility for robots should remain with people.
Why read it?
Offers an explicit position in the debate about responsibility for technical systems.
Keep in mind
Bryson defends a moral position. Read the title in the context of the argument.
Reference details
Title as recorded: Robots should be slaves
Author(s): Joanna J. Bryson · Year: 2010
Publication / host: John Benjamins Publishing Company
Source language: English
DOI: 10.1075/nlp.8.11bry
KP-R-D04BCFDD3D77
No sources match these filters. Try fewer filters or a different search term.