What Does 'Human-Centred AI' Mean?

2025/07/26 by Olivia Guest, Guest, Olivia · 9 voices · 2 citations
Neuroscience · Social Sciences · #cs.AI

paper · pdf · doi:10.48550/arxiv.2507.19960

Abstract

While it seems sensible that human-centred artificial intelligence (AI) means centring "human behaviour and experience," it cannot be any other way. AI, I argue, is usefully seen as a relationship between technology and humans where it appears that artifacts can perform, to a greater or lesser extent, human cognitive labour. This is evinced using examples that juxtapose technology with cognition, inter alia: abacus versus mental arithmetic; alarm clock versus knocker-upper; camera versus vision; and sweatshop versus tailor. Using novel definitions and analyses, sociotechnical relationships can be analysed into varying types of: displacement (harmful), enhancement (beneficial), and/or replacement (neutral) of human cognitive labour. Ultimately, all AI implicates human cognition; no matter what. Obfuscation of cognition in the AI context -- from clocks to artificial neural networks -- results in distortion, in slowing critical engagement, perverting cognitive science, and indeed in limiting our ability to truly centre humans and humanity in the engineering of AI systems. To even begin to de-fetishise AI, we must look the human-in-the-loop in the eyes.

Citations

• A metatheory of classical and modern connectionism.
• To Improve Literacy, Improve Equality in Education, Not Large Language Models
• Deflating Mental Representation
• Can We Trust AI Benchmarks? An Interdisciplinary Review of Current Issues in AI Evaluation
• Reclaiming AI as a Theoretical Tool for Cognitive Science
• ChatGPT is bullshit
• The Dark Side of Dataset Scaling: Evaluating Racial Classification in Multimodal Models
• What Makes a Good Theory, and How Do We Make a Theory Good?
• Into the LAIONs Den: Investigating Hate in Multimodal Datasets
• The Surveillance AI Pipeline
• On Hate Scaling Laws For Data-Swamps
• On Logical Inference over Brains, Behaviour, and Artificial Neural Networks
• On the Dangers of Stochastic Parrots
• Towards decolonising computational sciences
• Artificial Intelligence is stupid and causal reasoning won't fix it
• How Computational Modeling Can Force Theory Building in Psychological Science
• ELIZA—a computer program for the study of natural language communication between man and machine
• I.—COMPUTING MACHINERY AND INTELLIGENCE
• Models as Mediators

Cited by

• Brainrot: Deskilling and Addiction are Overlooked AI Risks
• Facial analysis AI as social pseudotechnology

Discussions

• Boiling here at home in Cyprus but I put the finishing touches a couple of days ago on this preprint: What Does 'Human-Centred AI' Mean? doi.org/10.48550/arX... Wherein I analyse HCAI & demonstrate t [bsky, 389 points, 11 comments]
• i really like this chart from @olivia.science's paper "What Does ‘Human-Centred AI’ Mean?" arxiv.org/pdf/2507.19960 [bsky, 29 points, 1 comments]
• It's important to read strong critiques of AI and @olivia.science writes some of the strongest. In this one, she re-centers the humans who create #GenAI systems & proposes a way to demystify AI, showi [bsky, 26 points, 2 comments]
• I guess you may not realize you are speaking to an expert in AI Can warmly recommend reading @olivia.science ‘s work arxiv.org/abs/2507.19960 [bsky, 17 points, 1 comments]
• displacement of human cognition, capitalist theft of labor, "AI" defined as a relationship between humans and technology, machine hauntology: I thoroughly recommend this insightful (and scorching) ana [bsky, 12 points, 1 comments]
• Wow! What a ranging, clear, and thoughtfully worded conceptual reengineering @olivia.science. doi.org/10.3390/bs16... [bsky, 10 points, 1 comments]
• What Does 'Human-Centred AI' Mean? arxiv.org/abs/2507.19960 interesting read from @olivia.science [bsky, 7 points, 1 comments]
• Para más detalles sobre las diferencias entre las tecnologías, y su impacto en los humanos, este artículo es denso pero muy ilustrativo y claro: arxiv.org/pdf/2507.19960 [bsky, 5 points, 1 comments]
• Yes, I was looking for the link to the paper on the types of AI. I am part of a software engineering researchers whatsapp group and they were discussing how to use AI in research arxiv.org/abs/2507.19 [bsky, 4 points, 2 comments]

Related

• A humanist perspective on artificial intelligence
• A.I. go by many names: towards a sociotechnical definition of artificial intelligence
• Human-Centered AI
• Human-Centered Artificial Intelligence and Machine Learning
• Understanding Human-Centred AI: a review of its defining elements and a research agenda