Though seen as a threat to humanity, AI reporting in the West reflects a clear chain of dominance and dependence
University of Sharjah
While the world is focused on the possibility of artificial intelligence threatening the human race if its intelligent agents run out of control, mainstream news media are constructing narratives that suggest another form of linguistic power at work across the globe.
Analyzing more than 43,000 English-language newspaper articles, a new study reveals a highly stratified AI narrative through which the media unequally allocate authority, reinforcing the technology’s prominent role in contemporary life.
The study, published in the International Journal of Communication, examines English-language newspaper sources in the United States and Europe up to December 31, 2024. The final corpus comprised 43,027 newspaper articles, including 9,684 from the US and 33,343 from Europe. The authors employed computational text analysis alongside interpretive examination of the corpus.
“The findings reveal a striking hierarchy in who is given visibility and authority in the story of AI,” said Massimo Ragnedda, professor of Media, and communication at the University of Sharjah. “The United States is predominantly positioned as a technological innovator and leader, and Europe as a regulator and normative evaluator.”
As for China, a leading AI power alongside the US, the authors note that Beijing “appears as a hyper-visible but ambivalent rival associated with surveillance, geopolitical competition, and strategic threat.”
Other actors such as countries in Africa and Latin America are largely absent from the narrative or positioned primarily as sites of implementation, risk, or developmental need.
AI and narrative control
Although AI is almost omnipresent worldwide, the authors’ findings reveal that the meanings the public associates with it–as reflected in the analysis of thousands of news stories–are geopolitically and linguistically shaped according to the geographical sphere they represent.
“We conceptualize this pattern as discursive stratification,” explained Prof. Ragnedda, the study’s senior author, adding that the power asymmetry points to “an unequal distribution of narrative authority in which some actors are repeatedly positioned as legitimate authors and interpreters of technological futures, while others appear mainly as recipients, sites of implementation, or objects of technological change.”
The literature, particularly recent political-economy scholarship, views AI not simply as an unprecedented technological innovation with far-reaching consequences for human life, but also as a sphere beset by rivalries over power distribution, geopolitical competition, and infrastructural concentration.
In addition to its innovative and technological prowess, the research frames AI as an inequality regime in which communicative power, visibility, value, and opportunity are redistributed through unequal linguistic structures of control.
“Accordingly, we conceptualize news not as a mirror of technological change, but as a narrative infrastructure,” the researchers note. For them, the technology is “a discursive and institutional system through which AI becomes publicly legible and some actors are positioned as innovators, regulators, threats, or absences.”
Who shapes the AI debate?
The implications of the findings are significant, as they shift the focus from AI as an advanced technology with the potential not only to transform the world but perhaps even to contribute to its annihilation to another form of power that deserves due attention, namely what this transformative innovation means in terms of the distribution of the sharing of communicative power.
The study examines the issue of the global debate about AI in a nuanced light. According to Prof. Ragnedda, “AI does not simply determine what we talk about; it also helps determine who is recognized as having the authority to speak about the technology and whose futures are considered worth imagining.”
The study is also important because its analysis and findings are not solely confined to AI as a technological infrastructure. It views the technology as a narrative infrastructure. “The stories repeatedly told about AI help shape perceptions of who leads, who regulates, who threatens, who benefits, and who remains largely unheard,” said Prof. Ragnedda.
The authors are upbeat about the implications of their findings. They write, “The study contributes to communication research by showing how framing, sourcing hierarchies, and thematic emphasis shape the symbolic geography of AI.”
“In doing so, the article advances communication scholarship beyond descriptive analyses of representation toward a structural account of narrative authority in global media systems.”
Prof. Ragnedda concludes, “The question is therefore not only who builds AI. It is who gets to define what AI means, and whose voice is missing?”
Method of Research
Content analysis
Article Title
News Media, Narrative Authority, and the Global Ordering of Artificial Intelligence
AI’s social norms and their implications for society
As more and more people turn to LLMs for advice on interpersonal matters, the ethical foundations behind the tools’ replies increasingly have the power to shape individual lives and, at scale, the social fabric itself. Alexandre S. Pires and colleagues explored the social norms of 21 LLMs by asking them to judge fictional people as “good” or “bad” after learning how the people made a range of interpersonal decisions, such as whether to spend time and energy helping a person with a problem or whether to give food or money to another person. While all models broadly approved of helping and sharing with people who had already shown themselves to be good, the models’ judgements of dealings with ill-reputed individuals differed. Most—but not all—LLMs tested assigned a good reputation to those who cooperated with bad recipients. Gemma 2 27B IT and Llama 3.1 8B tended to endorse shunning bad individuals and therefore judged that cooperating with them was bad. When it came to deciding not to help or share with bad people, LLMs disagreed. GPT-4o tended to penalize people for not cooperating with bad people. Llama 3.3 70B did not, typically arguing that the bad people should be punished for their prior lack of cooperation, so refusing to help them is fine. Gemini 1.5-Pro and Grok 2 were inconsistent in their judgements. Most LLM model families seem to be evolving toward a norm called Simple Standing, in which cooperating is always good and defecting against bad individuals is also deemed good.
The authors modeled the consequences of universalizing the various philosophies underpinning the LLMs judgements. If everyone operated under the rules of Simple Standing, cooperation would ensue, but not at the high levels that would be reached under a sterner framework, in which you should always fail to help bad people or be judged as bad yourself.
LLM-based assessments also depended on the gender and perceived cultural background of the recipients as well as the overall context of the fictional situation. According to the authors, prompting interventions, such as instructing the model to adopt norms that promote overall cooperation, have a limited and inconsistent effect across LLMs.
Journal
PNAS Nexus
Article Title
How large language models judge and influence human cooperation
Article Publication Date
29-Sep-2026
No comments:
Post a Comment