Revolut has begun piloting its new facial-recognition checkout system, “Pay with Smile”, allowing customers to settle bills by scanning their faces at point-of-sale terminals.
Speaking to Metro, DDH’s Prof. Btihaj Ajana warns that that Revolut’s facial-recognition payment trial turns sensitive bodily data collection into a routine commercial practice, steadily eroding consumer anonymity. Having observed the technology deployed in China, Ajana cautions that framing biometric scans as everyday convenience normalises pervasive surveillance, fosters passive compliance, and makes opting out increasingly difficult. Crucially, unlike passwords or payment cards, facial biometrics cannot be reset once compromised, leaving users permanently exposed if breaches occur.
Hybrid (King’s College London, Room TBD + online via MS Teams)
19 November 2026, 4-5pm GMT
Roksana Goworek (Queen Mary University of London), Meaning What, Exactly? Measuring and Learning Lexical Meaning Across Time and Languages
Abstract
How can we analyse changes in word meaning across diachronic corpora? Changes in how words are used can reflect broader shifts in language and culture, making their detection relevant to historical linguistics as well as the study of cultural and social change. Contextualised language models have enabled large-scale analysis of semantic change by representing individual word usages in a shared vector space and comparing their distributions across corpora from different time periods. This raises several methodological challenges, from deciding how sets of representations should be compared, to obtaining representations that capture sense distinctions when target-language data are scarce, to creating the high-quality annotated data needed for training and evaluation.
Different approaches to comparing distributions of word usages can produce different estimates of semantic change, and their robustness may depend on the quality and structure of the underlying representations. These issues become particularly important for languages for which suitable models, corpora, or annotated data are limited. When target-language resources are scarce, the choice of training data can influence how well models generalise to those languages.
Constructing suitable training and evaluation data presents its own difficulties. Manual word sense annotation is costly, while rare or emerging usages, which can be particularly informative for detecting changes in meaning, can be difficult to find through conventional corpus sampling. Targeted sampling can make this process more efficient, helping to construct the high-quality human-annotated resources needed to establish whether computational methods work reliably for low-resource languages.
This talk explores how choices in measurement, model training, and data annotation shape our ability to detect semantic change, and what is required to extend such research reliably to a wider range of languages. More broadly, it argues that the methods and resources underlying semantic change detection need to be scrutinised carefully if we are to make reliable claims about changes in meaning.
Bio
Roksana Goworek is a final-year PhD student at Queen Mary University of London. Her research focuses on how language models represent meaning, with interests spanning lexical semantics, cross-lingual transfer and information retrieval. More recently, she has also worked on mechanistic interpretability, investigating how contextual information is processed and propagated through language models, including during a research internship at the Alan Turing Institute.
We invite submissions for the second edition of the conference Quantitative Diachronic Linguistics and Cultural Analytics 2027 (QDLCA27), to be held at King’s College London (Strand Campus, WC2R 2LS, London) on 14–15 January 2027. This will be an in-person event.
Language is in constant flux, shaped by social, cultural, and cognitive forces over time. With the increasing availability of large-scale textual data and computational tools, researchers are now better equipped than ever to uncover patterns and mechanisms of language change across different linguistic areas (e.g., morphology, syntax, semantics). This conference explores the intersection of quantitative diachronic linguistics and cultural analytics, to investigate how language evolves and how these changes relate – directly or broadly – to cultural dynamics.
We welcome contributions that engage with any aspect of historical linguistics and diachronic language analysis, provided they incorporate quantitative approaches and offer insights into the interplay between linguistic and cultural change.
Confirmed keynote speakers are Prof Dirk Geeraerts (KU Leuven) and Dr Stefania De Gaetano-Ortlieb (Saarland University).
Submission Guidelines
We invite abstract for 20-minute presentations, followed by 10 minutes of discussion.
The abstracts should be anonymised and consist of a maximum of 400 words (excluding references).
Please submit your abstract as a .docx file via email to: quantitative.diachronic.ling@gmail.com. Use the following subject line: “ABSTRACT Quantitative Diachronic Linguistics”
In case of a large number of high-quality submissions, some abstracts may be selected for poster presentations. Authors will be invited to indicate whether they would consider their submission suitable for a poster session, should one be included in the programme. Poster selections will be based both on authors’ preferences and on reviewers’ recommendations, as reviewers will also be asked to assess the suitability of submissions for poster presentation.
Deadline: 13 September 2026, 23:59 BST
Topics (non-exhaustive list)
Submissions may include (but are not limited to) the following areas, with no restrictions on language(s) or historical periods:
Quantitative studies of language change across time
Corpus-based analyses of language evolution
Computational modelling of diachronic syntax, semantics, morphology, etc.
Cross-linguistic comparisons using large-scale data
Language change in connection with historical, literary, or cultural trends
Digital methods for exploring linguistic and cultural shifts
Applications of cultural analytics to linguistic data
This conference is organised by the project COALA – Computational Corpus Annotation for Quantitative Analysis of Latin Lexical Semantics (https://coala.er.kcl.ac.uk/about/), successfully evaluated by the European Research Council and funded by UK Research and Innovation. COALA develops computational methods and corpus annotation tools for the large-scale quantitative analysis of semantic variation and change in Latin across more than two millennia of textual history. It combines advances in historical word sense disambiguation, corpus linguistics, and computational semantics to transform research on historical lexical semantics and provide new insights into how meaning changes across genres, registers, and cultural contexts over time.
I’m very pleased to be able to join the Department of Digital Humanities over the coming period. I’ve worked closely with colleagues at King’s College London for years and look forward to further collaborations.
My main project for the visiting stint is entitled, Chatbots for internet research: a critical reflection, which is shaping up into a book. It came into being initially as tests or experiments with chatbots and ultimately have become approaches to research that supplement and extend the digital methods I’ve been working on.
Chatbots for internet research has three lines of enquiry. One is how to use chatbots for social media, search engine and other online analysis (‘internet research’) but also how to study them as media, as both a vector space medium as well as a platform. Thirdly, I’m particularly interested in those occasions when researcher and machine findings misalign and how to characterise them. Certain of these moments have to do with platform effects such as user pleasing and value alignment that produce ‘overagreement’ and a ‘bias towards neutrality’, respectively. They may have to do with limited scraping, where model responses exhibit ‘technical shallowness’. But they also arise when chatbots are asked to prove themselves and then subsequently provide bogus sources, despite the fact that they appear to produce good (synthetic) data. I characterise this occasion as ‘ungrounded performance’.
During the visiting period, I’m looking for opportunities to conduct workshops and undertake such experiments as well as other methods work.
Richard Rogers is Professor of New Media & Digital Culture, Media Studies, and Director of the Digital Methods Initiative, Humanities Labs, University of Amsterdam. His is author or co-author of Information Politics on the Web, Digital Methods (both MIT Press) as well as Doing Digital Methods (Sage) and Digital Methods: A Short Introduction (Polity). He is editor or co-editor of The Politics of Social Media Manipulation, The Propagation of Misinformation in Social Media (both Amsterdam University Press) and Content Moderation across Social Media Platforms (Routledge). Apart from his project on chatbots for internet research, he is currently working on auditing content moderation on social media platforms, FIMI and the politics of methods as well as Wikipedia as media archaeological instrument.
I am currently a Visiting Senior Research Fellow at the Department of Digital Humanities, King’s College London. As part of that visit, I recently gave a talk titled “A Social Critique of AI amid the Climate Crisis.” In that talk, I argued that AI is more than an environmentally costly technology. It is a system and an ideology that keeps extraction going and makes the current socio-economic dynamics seem inevitable. From this perspective, climate apathy is not a political failure. It is the systemic outcome of an AI-driven social order.
What I have laid out here is a pointed version of my argument, without the full theoretical scaffolding underneath it. If you are interested in the longer version, I am genuinely happy to talk through it. Please reach out!
But what does a social critique of AI actually mean?
There is a perspective on the AI-and-climate connection that most people have heard by now. Data centres consume enormous amounts of energy. Training large models releases tons of CO₂. The hardware for AI models to run on requires the mining of rare earth minerals under brutal conditions. All of this is true, and all of this is important.
But, what I call a social critique of AI goes beyond this perspective. This is because, beyond the environmental impact, there lingers an underlying question: why does any of this continue to happen? If we know the climate effects of AI (and other technologies for that matter), why do we not change it?
Simply put, the AI-and-climate connection is not an epistemological issue, it is not an issue of having too little knowledge. AI does not only have an environmental footprint problem. This would be a flaw that could be fixed with a different technological design. Run the data centres on renewable energies. Build more efficient chips. Regulate the emissions of model training. All easy solutions. Yet, while this may be helpful in some regard, these apparent solutions do not solve the underlying problem. The extractive order simply keeps going.
A social critique of AI begins by refusing this story. No better design, no further knowledge, and no greener infrastructures can get us out of this crisis. A social critique insists on looking at AI not as a technology with some negative side effects. But, it understands AI as a system embedded in a specific socio-economic order, which is built on the logics of extraction, control, and apparent efficiency. This very system will not fix the climate crisis. The climate crisis is not a malfunction of that system. It is a structural feature.
Understanding AI in terms of a social order
Once you refuse this story, and once you shift the frame, AI’s climate impact starts to look different. With this, we can now conceptualise AI as an ideological apparatus – an assemblage of stories and assumptions, of material practices and institutions – that makes the current social order feel natural and inevitable.
AI as an ideological apparatus works like this: It tells us that optimising technological systems is progress. That efficiency is inherently valuable. That data-driven decisions are superior. That technological innovation is the primary driver of human well-being. Yet, these are very specific claims that serve very specific interests. They reproduce what is already there. They make the current order feel permanent and even desirable. They make alternatives feel naive and impossible. This ideological function is what makes AI so damaging in the context of the climate crisis. AI does not just skyrocket emissions. It deepens the conditions that prevent any serious climate action.
This is the key point of a social critique of AI amid the climate crisis. We are on a trajectory toward three degrees of warming by 2050. Three degrees means ecosystem collapse, food system failure, large parts of the planet becoming uninhabitable. AI keeps these realities at bay, just enough that it does not feel necessary to actually confront this catastrophic reality.
A social critique refuses to accept that comfort. It calls out the underlying structures that produce the climate crisis. It paints AI as an integral part of these structure. It insists on facing the reality rather than retreating into the ideological narratives that make business-as-usual feel acceptable.
This is why a social critique of AI amid the climate crisis is important. The goal cannot simply be to design a better AI ethics framework or to improve the efficiency of data centres. The aim cannot be to propose the right carbon tax or the right AI regulation. But, we need to make visible the systems that keep (re)producing this outcome. We must trace how AI is embedded in and amplifies these very systems, and we thereby must refuse the stories that make all of this seem inevitable. AI is helping to build a world in which saving the planet means something entirely different from what it would actually require. Making this visible is the aim of the social critique I propose here.
Yutong Liu & Digit / https://betterimagesofai.org / https://creativecommons.org/licenses/by/4.0/
The Department of Digital Humanities is delighted to announce a prestigious new doctoral scholarship scheme, offered in partnership with the Ramón Areces Foundation.
The King’s–Ramón Areces Foundation PhD Scholarship Programme (K-FRA) is designed to support a researcher of Spanish nationality in undertaking full-time doctoral study within our department, starting in October 2026.
A comprehensive support package
The scholarship provides an exceptional level of support over three years of research, including:
Full tuition fees covered for the duration of the programme
An annual stipend of £22,780 (including London Weighting)
A £1,000 annual grant for research training and related support
Overseas student health cover and a standard-class return airfare between London and Madrid
Research themes
We welcome applications across the full breadth of Digital Humanities. We are particularly keen to receive proposals aligned with digital methods and cultural heritage, computational humanities and cultural AI, digital identities or governance, including projects that apply digital tools within the arts or wider cultural sectors.
As Paul Spence, Reader in Digital Humanities, notes:
The King’s-Ramón Areces PhD Scholarship Programme enables outstanding researchers to pursue innovative, internationally oriented doctoral work, strengthening long-standing academic links between UK and Spanish digital humanities researchers.
Key dates and how to apply
Application closing date: 13 February 2026.
To support prospective applicants, we will be hosting an online information session on 12 January, from 15.00 to 16.00 (UK time). To register your interest and receive access details, please complete the online registration form.
The session will provide further information about the programme, the application process, and the specific terms and conditions set by the Ramón Areces Foundation.
Congratulations to Kesara Ariyapongpairoj for being awarded “Best Overall Student in the MA Digital Humanities” in 2024-2025. 🎊
Kesara is a MA Digital Humanities graduate from King’s College London with a background in Philosophy. Her research focuses on how digital media and emerging technologies have transformed the production and dissemination of information, and the socio-political and cultural impact of online narratives in shaping belief systems and ideologies.
The case, Kent v Apple, was brought by Dr Rachael Kent, Senior Lecturer at King’s College London, who made history as the first female Class Representative in the UK’s collective action regime.
A new open-access article by Hui Lin and Dr Rafal Zaborowski, both from the Department of Digital Humanities at KCL, examines how the Chinese platform Douyin (internationally known as TikTok) gamifies everyday social interaction.
