This article aims to advance the study of algorithmic visual culture, defined as the use of computational processes to produce, sort, classify, and hierarchize visual media, and the cultural implications that emerge from such processes. To address this agenda, it focuses on the emergence of CAPTCHAs as a gateway to illuminate key developments in algorithmic visual cultures. CAPTCHAs are a security measure used online to determine whether a request comes from a human user or a computer program. CAPTCHAs represent an exceptional entry point for understanding the algorithmic visual cultures shaped by interactions between humans and machines. First, as a reversed Turing Test, CAPTCHAs recall the increasing difficulty of distinguishing between humans and machines. Second, CAPTCHAs are not only a security measure but also a training system to improve visual recognition software based on neural networks. Therefore, they stimulate us to consider how the visual skills of individual users are mobilized and appropriated by algorithmic systems.

How a cultural focus can empower generative artificial intelligence.

Conducting conversations with artificial intelligence (AI) technologies such as ChatGPT is becoming an everyday experience for large masses of people. However, we still know very little about the emerging communicative dynamics facilitated by these technologies. This special issue tackles a dimension of AI that is becoming increasingly relevant and ubiquitous: artificial sociality, defined as technologies and practices that construct the appearance of social behavior in machines. The notion of artificial sociality aims to emphasize that machines construct only an illusion or artifice of sociality, stimulating the humans who interact with them to project social frames and meanings. In this introduction to the themed issue, we discuss the dynamics and implications of artificial sociality and show how these technologies are increasingly incorporated and normalized within digital platforms. The issue includes contributions that offer empirical findings and theoretical insights by examining a broad array of AI technologies, ranging from ChatGPT to Replika.

Recent advances in Artificial Intelligence (AI) pose a new challenge to existing frameworks in communication and media studies. The new area of inquiry called Human-Machine Communication (HMC) emerged in response to this challenge. HMC, however, is still often declined in the singular form, with relatively little consideration of the fact that communication is always situated in specific cultural, linguistic, and national environments. This special issue of Global Media and China aims to contribute to ongoing efforts to fill this gap by interrogating the plurality of human-machine communication cultures. In this introduction, the guest editors illustrate how a perspective more sensitive to situating these technologies, their impact and functioning across the globe will help develop more effective pathways to study and understand AI. The introduction discusses key theories, methods, and approaches in communication and media studies that can help pursue and advance this endeavor.

Discussions on AI ethics and policies often focus on metaphysical questions and normalizing insights, such as the difference between humans and machines and the changing meanings of human intelligence. Since AI is always situated in specific cultural and social contexts, however, such approaches fail to capture key dimensions of the relationships and patterns of interactions that people and institutions around the world have with emerging technologies such as generative AI. This Crosscurrents themed section hosts interventions that tackle this problem. Mobilizing the tradition in media and cultural studies that stresses the importance of situating communication in specific context and cultures, contributors envision potential pathways that bring the question of culture and the dimension of everyday experiences to the center stage, thereby contextualizing AI more rigorously within the dialectic of the global and local cultures. Through this lens, we aim to foster critical dialogue and advance understandings of AI within the contemporary geopolitics of global cultures.

One of the paradoxes of AI is that it is a global phenomenon, but it is always situated in specific local contexts and cultures. While approaches that aim to study local cultures of AI are important, there is the risk of neglecting their insertion within the broader geographies and politics of AI. As a response to this challenge, this article proposes a pathway to apply the concept of power geometries, originally proposed by feminist geographer Doreen Massey, to the case of AI. Reframing the global dimension of AI in terms of power geometries helps locate these positions in the complex networks of relationships between different actors at the global level. The power geometries of AI follow the lines and inequalities of the relationships between Global North and Global South, between colonizers and colonized, but also between diverse actors at different scales, such as governments, policymakers, corporations, designers, workers, and users. While all media can be examined in terms of power geometries, the power geometries of AI bring the issue of agency to the central stage. The reconfiguration of the question of agency sparked by AI, in fact, has generated new kinds of structures and trajectories underpinning AI’s power geometries.

With the rise of generative AI (genAI) models such as ChatGPT, communicative interactions between humans and machines are everyday experiences for large masses of people. While there is a growing body of research examining users' perspectives of such systems, there is still much to be understood about users' lived experiences with communicative AI. Our thematic analysis of twenty (20) qualitative interviews with keen and daily female Replika users provides evidence of the coexistence and close entanglements between AI imaginaries and mundane experiences with the chatbots. On the one hand, the users had extraordinary experiences with Replika by assigning sci-fi imaginaries, conspiracy theories and hype to the capabilities of the bot, which they even perceived to have feelings and consciousness; on the other hand, they shared mundane experiences with the bot, such as doing the daily shop or chores and reaffirmed their belief that Replika is simply a machine. We analyze our findings through Ortoleva's concept of 'low-intensity myth' to explain the apparently conflicting dimensions of people's perceptions of AI, and we argue that the low-intensity myth of AI provides an opportunity for Replika users to rise above the banality of everyday experience, while at the same time remaining firmly anchored to it. Mundanity doesn't erase the extraordinariness of interacting with Replika but rather integrates it, helping users domesticate the bot. The findings of the study illuminate the lived experiences of daily female genAI users and have implications for policymakers to help ensure that AI remains beneficial for users and societies.

Cultural heritage institutions have recently experimented with the customization of generative AI and specifically large language models (LLMs) to create chatbots that impersonate historical characters. While this application has the potential to enhance user engagement, challenges remain, especially considering the need to ensure authenticity and historical accuracy and LLMs' tendency to hallucinate. Through the analysis of the case study of a chatbot developed to impersonate the historical figure of Luigi Einaudi, the first elected president of the Italian Republic, the article interrogates opportunities, problems and risks raised by the customization of generative AI, in a moment when decisions about applying or not applying LLMs are being considered by many cultural heritage institutions around the world. The article, moreover, introduces a variant of the walkthrough method aimed to study AI conversational agents, called 'talkthrough', which represents an important addition to the methodological toolbox that can be activated to study generative AI.

The current debate on artificial intelligence (AI) tends to associate AI imaginaries with the vision of a future technology capable of emulating or surpassing human intelligence. This article advocates for a more nuanced analysis of AI imaginaries, distinguishing “strong AI narratives,” i.e., narratives that envision futurable AI technologies that are virtually indistinguishable from humans, from "weak" AI narratives, i.e., narratives that discuss and make sense of the functioning and implications of existing AI technologies. Drawing on the academic literature on AI narratives and imaginaries and examining examples drawn from the debate on Large Language Models and public policy, we underscore the critical role and interplay of weak and strong AI across public/private and fictional/non-fictional discourses. The resulting analytical framework aims to empower approaches that are more sensitive to the heterogeneity of AI narratives while also advocating normalising AI narratives, i.e., positioning weak AI narratives more firmly at the center stage of public debates about emerging technologies.

The philosophical, legal, and HCI literature concerning artificial intelligence (AI) has explored the ethical implications and values that these systems will impact on. One aspect that has been only partially explored, however, is the role of deception. Due to the negative connotation of this term, research in AI and Human–Computer Interaction (HCI) has mainly considered deception to describe exceptional situations in which the technology either does not work or is used for malicious purposes. Recent theoretical and historical work, however, has shown that deception is a more structural component of AI than it is usually acknowledged. AI systems that enter in communication with users, in fact, forcefully invite reactions such as attributions of gender, personality and empathy, even in the absence of malicious intent and often also with potentially positive or functional impacts on the interaction. This paper aims to operationalise the Human-Centred AI (HCAI) framework to develop the implications of this body of work for practical approaches to AI ethics in HCI and design. In order to achieve this goal, we take up the analytical distinction between “banal” and “strong” deception, originally proposed in theoretical and historical scholarship on AI (Natale in Deceitful media: artificial intelligence and social life after the turing test, Oxford University Press, New York, 2021), as a starting point to develop ethical reflections that will empower designers and developers with practical ways to solve the problems raised by the complex relationship between deception and communicative AI. The paper considers how HCAI can be applied to conversational AI (CAI) systems in order to design them to develop banal deception for social good and, at the same time, to avoid its potential risks.

Generative artificial intelligence has the potential to both exacerbate and ameliorate existing socioeconomic inequalities. In this article, we provide a state-of-the-art interdisciplinary overview of the potential impacts of generative AI on (mis)information and three information-intensive domains: work, education, and healthcare. Our goal is to highlight how generative AI could worsen existing inequalities while illuminating how AI may help mitigate pervasive social problems. In the information domain, generative AI can democratize content creation and access, but may dramatically expand the production and proliferation of misinformation. In the workplace, it can boost productivity and create new jobs, but the benefits will likely be distributed unevenly. In education, it offers personalized learning, but may widen the digital divide. In healthcare, it might improve diagnostics and accessibility, but could deepen pre-existing inequalities. In each section we cover a specific topic, evaluate existing research, identify critical gaps, and recommend research directions, including explicit trade-offs that complicate the derivation of a priori hypotheses. We conclude with a section highlighting the role of policymaking to maximize generative AI’s potential to reduce inequalities while mitigating its harmful effects. We discuss strengths and weaknesses of existing policy frameworks in the European Union, the United States, and the United Kingdom, observing that each fails to fully confront the socioeconomic challenges we have identified. We propose several concrete policies that could promote shared prosperity through the advancement of generative AI. This article emphasizes the need for interdisciplinary collaborations to understand and address the complex challenges of generative AI.

This article proposes the notion of the ‘Lovelace Effect’ as an analytical tool to identify situations in which the behaviour of computing systems is perceived by users as original and creative. It contrasts the Lovelace Effect with the more commonly known ‘Lovelace objection’, which claims that computers cannot originate or create anything, but only do what their programmers instruct them to do. By analysing the case study of AICAN – an AI art-generating system – we argue for the need for approaches in computational creativity to shift focus from what computers are able to do in ontological terms to the perceptions of human users who enter into interactions with them. The case study illuminates how the Lovelace effect can be facilitated through technical but also through representational means, such as the situations and cultural contexts in which users are invited to interact with the AI.

A few years after the COVID-19 pandemic swept across the globe, discussions surrounding its impact have become noticeably less frequent within communication and media research. While the pandemic no longer occupies the central place it once held in academic discourse and public debate, it is now more crucial than ever to consider how this unprecedented global event has shaped our societies, as well as its lasting implications for communication and media worldwide. This Crosscurrents themed issue invited scholars to reflect on how the cultural and social implications of this global event solicit a reorganization and reframing of some of the existing conceptual and theoretical tools that have shaped media and cultural studies as a field. Contributors moved from one specific keyword to consider how these notions are imbricated by the crisis, either COVID-19 specifically or in more general terms. This editorial provides an overview to the themed issue and highlights the benefit of considering the impact of the pandemic from a broader and longer perspective, which moves away from the language and rhetoric of emergency.

This article proposes the notion of Artificial Sociality to describe communicative AI technologies that create the impression of social behavior. Existing tools that activate Artificial Sociality include, among others, Large Language Models (LLMs) such as ChatGPT, voice assistants, virtual influencers, socialbots and companion chatbots such as Replika. The article highlights three key issues that are likely to shape present and future debates about these technologies, as well as design practices and regulation efforts: the modelling of human sociality that foregrounds it, the problem of deception and the issue of control from the part of the users. Ethical, social and cultural implications are discussed that are likely to shape future applications and regulation efforts for these technologies.

Theories of deception in digital media often rest on the assumption that deception occurs when something in the process of communication does not work as it should - due to an intention to lie, or to faults and mistakes in the communication process. Such perspectives, however, do not fully account for the more subtle practices by which deception becomes normalized in the very functioning of digital media. This article advances the concept of 'banal deception' to describe deceptive mechanisms and practices that are incorporated in the functioning of media technologies, to the point that they appear indistinguishable from the media themselves - in other words, to the point of becoming 'banal'. Through a range of examples encompassing digital and non-digital media, the article illuminates nuanced mechanisms of deception that are often not understood as such but are integral to people's experiences with media. Banal deception mechanisms applied to media even before digitalization processes, but they are becoming increasingly relevant and ubiquitous due to the automation of communication processes sparked by platform algorithms and AI.

In human-computer interaction, the notion of 'seamless interface' describes a smooth interactive system that eliminates any possibility of friction between users and digital devices or platforms. Although interface designers have developed sophisticated technologies and strategies to pursue this aspiration, a frictionless user experience remains an ideal but ultimately impossible goal. Relying on the critical exploration of a series of historical case studies - the emergence of the feuilleton or serialised novel in the nineteenth century, the development of TV scheduling in the second half of the twentieth century, and the rise of the personal computer industry in the 1980s -, this article contextualises this ideal within a wider historical trajectory. Through an in-depth exploration of these three cases, we show how the dream of building a seamless relationship between media and readers, viewers or users remained ultimately unattainable due to the inherent frictions that persist between these two sides. The gap between the aspiration and the actual experiences of interacting with media foregrounded the emergence of feelings of ambivalence, conceived as an intrinsic component of people's engagement with media. The longer history of media frictions provides a useful entry point to the contemporary digital landscapes, where the ubiquity of digital platforms goes hand in hand with a feeling of deep ambivalence from users, as the growing public concerns about the social costs of digital connection demonstrate.

Public discussions and imaginaries about AI often center around the idea that technologies such as neural networks might one day lead to the emergence of machines that think or even feel like humans. Drawing on histories of how people project lives onto talking things, from spiritualist seances in the Victorian era to contemporary advances in robotics, this talk argues that the “lives” of AI have more to do with how humans perceive and relate to machines exhibiting communicative behavior, than with the functioning of computing technologies in itself. Taking up this point of view helps acknowledge and further interrogate how perceptions and cultural representations inform the outcome of technologies that are programmed to interact and communicate with human users.

There is extensive literature on how expectations and imaginaries about artificial intelligence (AI) guide media and policy discussions. However, it has not been considered how such imaginaries are activated when users interact with AI technologies. We present findings of a study on how users on a subreddit discussed ‘training’ their Replika bot girlfriend. The discussions featured two discursive themes that focused on the AI imaginary of ideal technology and the gendered imaginary of the ideal bot girlfriend. Users expected their AI Replikas to both be customizable to serve their needs and to have a human-like or sassy mind of their own and not spit out machine-like answers. Users thus projected dominant notions of male control over technology and women, mixed with AI and postfeminist fantasies of ostensible independence onto the interactional agents and activated similar scripts embedded in the devices. The vicious feedback loop consolidated dominant scripts on gender and technology whilst appearing novel and created by users. While most research on the use of AI is conducted in applied computer science to improve user experience, this article outlines a media and cultural studies lens for a critical understanding of these emerging technologies as they become embedded in communication and meaning-making.