Men dominate AI media coverage in Spain

A significant disparity in media coverage of artificial intelligence (AI) has been uncovered in Spain, revealing a striking imbalance in the representation of men and women in news stories related to AI. According to a recent report, men are disproportionately featured as main subjects in 41% of AI-related news reports, while women account for just 5%.

The study, conducted by researchers at the University Institute of Gender Studies at Universidad Carlos III de Madrid (UC3M), analyzed a vast corpus of data consisting of 7,115 news stories, 13,787 journalistic sources, and 11,266 photographs and images published between 2017 and 2021, as well as in 2024. This extensive dataset was manually gathered by approximately 20 researchers who are also developing an algorithm to automatically detect biases in media coverage.

The report highlights a stark gender divide in the focus of media attention on individuals appearing in AI-related news stories during the analyzed period. Specifically, the distribution of media focus is characterized by a significant skew towards men, with 41.60% of reports focusing on male subjects, compared to only 5.33% that feature women.

The findings suggest that there may be an underlying "glass ceiling" limiting the representation of women in AI-related news stories in Spanish media. Further analysis and investigation are needed to fully understand the implications of these results and to identify potential strategies for promoting greater gender balance in AI media coverage.

The study's analysis of proper names in headlines reveals a stark imbalance, with men's names featured in 13.11% of news stories, while women's names appear in just 4.19%. This disparity underscores the limited visibility given to women in AI media coverage.

Men dominate the selection of journalistic sources, accounting for 61.14% of quoted individuals compared to 21.45% for women. The gap is even more pronounced when considering C-suite or presidential positions, where men are interviewed at a rate that far exceeds their female counterparts.

A closer examination of bylines shows that stories written by female journalists tend to include a higher proportion of female sources, with 25.58% of such articles featuring women compared to 18.96% in those written by male journalists. However, male bylines still hold sway in newsrooms, comprising 49.02% of all bylined content.

The increasing reliance on corporate and institutional sources has further exacerbated the issue, with this category accounting for 29.90% of sources in 2024. As a result, space for independent female experts is being squeezed out of the narrative.

Narratives around technical development, corporate strategy, and AI economics are largely dominated by male or institutional voices, leaving women's perspectives relegated to discussions on the social, ethical, educational, or cultural impact of technology, where they represent 29.26% of sources.

The study highlights a persistent issue in AI media coverage in Spain: sexist language continues to be prevalent despite overall improvements.

One notable bias is the use of generic masculine pronouns, which can perpetuate the assumption that experts are male unless explicitly stated otherwise.

This linguistic shortcut can lead to invisibility biases when female researchers and women in general are overlooked or relegated to secondary roles. The study notes that news stories often default to using the generic masculine when referring to professionals or award winners.

The data suggests that sexist language is still a significant issue, with 83.12% of articles featuring women incorporating some form of bias. Furthermore, 81.06% of these articles rely on the generic masculine pronoun.

In addition to linguistic biases, the study also examines visual representation in media coverage, revealing striking disparities in how men and women are portrayed in relation to AI technology.

The study's findings on visual representation in media coverage are equally striking. Women are often depicted as assistants or users of AI technology, while men are portrayed as developers, innovators, and leaders. This disparity is evident in the way women are frequently shown interacting with robots in a domestic setting, whereas men are depicted in more professional environments. The distinction highlights the limited roles and expectations assigned to women in relation to AI.

To address these biases, researchers have developed guidelines for journalists on how to promote more inclusive coverage of AI. One key recommendation is to diversify sources by including women in executive positions and technical development. This can be achieved through gender-balanced contact lists that provide visibility to female professionals who are often overlooked or underrepresented in media reports.

The guide also suggests abandoning generic masculine language, which perpetuates the notion that men are the default users of technology. Furthermore, it advises against objectifying robotic depictions by sexualizing them or reducing them to mere assistants. By implementing these changes, journalists can work towards a more nuanced and accurate representation of women's roles in AI development.

The reversal rule is another tool proposed for reviewing content. This involves swapping female and male gender terms to test whether the change affects sentence perception or reveals underlying stereotypes. Researchers note that applying this test can help identify unequal treatment and revise language to avoid sexist expressions, ultimately promoting a more inclusive and accurate representation of women in AI coverage.

The study reveals a stark disparity in how women and men are represented in AI-related media coverage in Spain. This unequal treatment is evident in the language used to describe individuals, with women often being referred to by their first names while men are addressed by their titles.

The researchers highlight that this biased representation can have far-reaching implications, perpetuating existing social inequalities and reinforcing stereotypes. By acknowledging these disparities, media outlets can take steps to revise their language and promote a more inclusive and accurate portrayal of women in AI coverage. Ultimately, this shift will contribute to a fairer and more nuanced understanding of the field's significance for all individuals.