A few days ago, one of Armenia\u2019s leading media outlets reported on a case of domestic abuse under the headline: \u2018Beaten out of jealousy?\u2019<\/p>\n
Within hours, the same narrative spread rapidly across Armenia\u2019s media landscape. <\/p>\n
Following a wave of public backlash, the original publisher amended the headline on its website. Yet on YouTube<\/a> and social media, the original framing remained untouched.<\/p>\n The episode was a stark reminder that the debate over gender-sensitive journalism is far from over. But today, the stakes are vastly higher than they used to be. <\/p>\n This is no longer just a question of how language shapes contemporary public opinion. Today, that exact same language is becoming part of AI\u2019s training data.<\/p>\n When jealousy takes centre stage<\/p><\/blockquote>\n \u2018A crime of passion,\u2019 \u2018He just couldn\u2019t accept the break-up\u2019 and \u2018She wasn\u2019t exactly a saint\u2019: the wording may vary from country to country, but the underlying pattern is universal.<\/p>\n For decades, news coverage of domestic and sexual violence has relied on a language that subtly shifts attention from the violence itself and onto the emotional state of the perpetrator or the history of the victim.<\/p>\n The issue is not the word \u2018jealousy\u2019 itself, but rather what that word does to the framing of a story. When violence is framed as the natural byproduct of uncontainable passion, the crime is pushed into the background. The audience is invited to empathise with, or at least analyse, the perpetrator\u2019s emotional distress.<\/p>\n Similarly, when reports on sexual assault linger on what a victim wore, how she behaved, or whether she had been drinking, the attention shifts from the offender to the target. <\/p>\n This does more than merely dilute accountability; it triggers victim-blaming \u2014 one of the oldest defence mechanisms in media history. In doing so, it obscures a much more uncomfortable reality: gender-based violence is rarely about love or passion. It is about power, control, and systemic inequality.<\/p>\n During a recent workshop organised by Women\u2019s Agenda<\/a>, I analysed these very patterns alongside journalists from various Armenian newsrooms. Despite their different editorial backgrounds, almost every group arrived at the same conclusion: phrases like \u2018he couldn\u2019t accept the split\u2019 unintentionally rationalise the crime, while fixating on a victim\u2019s clothing risks re-traumatising them in print.<\/p>\n These patterns are not confined to the crime beat. Even stories celebrating successful women frequently rely on headlines that filter their achievements through the lens of domestic stereotypes rather than professional merit.<\/p>\n This is precisely why gender-sensitive journalism is not a matter of superficial political correctness. It is about responsible editorial judgement, anchored in medicine\u2019s and media\u2019s oldest ethical code: do no harm<\/i>.<\/p>\n What are we feeding the machine?<\/p><\/blockquote>\n Over the past year of working with media professionals and journalism students, I have noticed a glaring blind spot. Most conversations regarding AI centre on a singular anxiety: How should journalists use tools like ChatGPT or Gemini?<\/i><\/p>\n We rarely stop to ask the inverse: What are these tools learning from us?<\/i><\/p>\n AI possesses no moral compass. It cannot differentiate between public-interest journalism and cheap sensationalism; it simply synthesises patterns from the data it consumes. <\/p>\n As the United Nations Development Programme (UNDP) has repeatedly warned<\/a>, when training data is steeped in bias, large language models do not correct it \u2014 they codify it.<\/p>\n Some researchers aptly describe these models as \u2018digital parrots<\/a>\u2019. They reproduce complex linguistic structures with astonishing fluency, entirely divorced from any actual understanding of what those words mean. AI did not invent our gender stereotypes; it merely inherited them.<\/p>\n Bias by design<\/p><\/blockquote>\n The same pattern appears in AI-generated images<\/a>. Ask an AI image generator to depict a surgeon, a lawyer or a chief executive, and the output will almost invariably skew heavily male.<\/p>\n This is hardly a technical glitch. Image-generation models learn from the collective visual culture we have spent decades archiving. <\/p>\n Editorial choices \u2014 the photographs selected, the illustrations commissioned, the visual framing employed \u2014 carry immense weight. And unlike the printed broadsheets of the past, those choices no longer disappear into tomorrow\u2019s recycling bin.<\/p>\n Today, a flawed headline might be scrubbed from a live article, but it lives on in YouTube algorithms, screenshots, and search indexes. Eventually, it is incorporated into future data used to train the next generation of language models. <\/p>\n A single careless editorial decision now possesses a digital half-life far longer than its creators ever anticipated.<\/p>\n Viewed through this lens, this is not fundamentally a story about technology. It is a story about the maturation of a trade. <\/p>\n Journalism has always evolved; practices deemed perfectly acceptable fifteen years ago no longer pass muster under modern professional standards. Acknowledging this is not about assigning historical blame, but about professional growth.<\/p>\n Perhaps that will be AI\u2019s greatest, if accidental, contribution to the industry: forcing us to re-examine the toxic editorial habits we long took for granted.<\/p>\n The human deficit<\/p><\/blockquote>\n The public debate remains obsessed with whether AI will ultimately replace journalists. The more urgent question is: What kind of journalism are we feeding it?<\/i><\/p>\n Generative AI writes with remarkable confidence, even when it is entirely hallucinating. It is incapable of self-doubt.<\/p>\n Journalists, however, must doubt. <\/p>\n Questioning a headline, interrogating a frame, challenging our own internal biases \u2014 this remains an exclusively human responsibility. Challenging the gender stereotypes embedded in our language is not an ideological luxury; it is a fundamental pillar of rigorous reporting.<\/p>\n Because if we continue to frame violence through the language of jealousy, sooner or later AI will do the same. Not because they believe it, but because we taught them no better.<\/p>\n","protected":false},"excerpt":{"rendered":" A few days ago, one of Armenia\u2019s leading media outlets reported on a case of domestic abuse under the headline: \u2018Beaten out of jealousy?\u2019 Within hours, the same narrative spread rapidly across Armenia\u2019s media landscape. Following a wave of public backlash, the original publisher amended the headline on its website. Yet on YouTube and social Read the full article…<\/a><\/p>\n","protected":false},"author":3,"featured_media":46604,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"ngg_post_thumbnail":0,"footnotes":""},"categories":[16,212],"tags":[],"class_list":["post-46635","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-critique","category-featured-post","author_posts-tatev-hovhannisyan"],"acf":[],"_links":{"self":[{"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/posts\/46635","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/comments?post=46635"}],"version-history":[{"count":1,"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/posts\/46635\/revisions"}],"predecessor-version":[{"id":46636,"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/posts\/46635\/revisions\/46636"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/media\/46604"}],"wp:attachment":[{"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/media?parent=46635"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/categories?post=46635"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/media.am\/en\/wp-json\/wp\/v2\/tags?post=46635"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}