Journalistic authenticity and the trade for ubiquity

Sometime around 1996 I was working as a systems administrator for a ‘New Media’ web design agency in Shoreditch, London. In a technological sense it was a time of rapid change; hardware was developing fast, and software likewise was making leaps; Tim Berners-Lee’s World Wide Web was developing from an environment of static links on flat pages to an interactive platform before my eyes. And yet I was somewhat blind to the consequences. One day I witnessed a group of developers discussing how to format web pages for mobile phones, and I just could not envision it happening. These people were pioneers of this new technology, making plans for a ‘future imaginary’ where content would not only be active but ubiquitous. I am a little ashamed to admit it but, even though I was immersed in the technology at the time, I lacked their vision of the future. Thirty years later I am witnessing another such moment. The rise of artificial intelligence agents has been meteoric in both its development and uptake, yet I suspect it is still in its infancy. Even at this stage I regularly witness people on public transport listening to, or watching, content that is obviously AI-generated and narrated by a synthetic voice, but people are engrossed. The fact that their content is generated by a large language model does not appear to bother them, if they are even aware of it. So is this the future we should expect, an era of undiscriminating consumption, and if it is, is it acceptable? Is the value of a piece of media output to be found in audience experience regardless of authorship?

In 'The Work of Art in the Age of Mechanical Reproduction' Walter Benjamin famously asks what happens to art when technology allows its reproduction and dissemination on a mass scale. There is a trade-off at work, he suggests: the uniqueness, history, authenticity, and 'aura' of an original piece are exchanged for accessibility, circulation, and reach. Yet AI complicates this equation because, although its output is based on the vast corpus of human work upon which it has been trained, the actual content produced lacks conscious creative intent, authenticity and, therefore, aura while at the same time providing abundance, accessibility, and specificity through personalised content. The equation therefore becomes unbalanced in that no original human artefact surrenders its prestige in return for ubiquity.

Perhaps, though, this is where a distinction can be found between fiction and journalism on one level at least. If a person seeks only temporary distraction during a bus journey, whether it matters that the story was produced by a person or a machine is moot. Yet to reduce all fiction to the experience of its consumption and ignore its potential value as a skilful expression of the human experience would be crass in the extreme. Human creativity is, surely, more than the mere production of consumable output: it is the skilful expression of subjective experience whether fact or fiction. Journalism therefore further complicates the issue, because, as a creative endeavour, its value rests upon a direct relationship to reality. A human journalist does not merely produce content but witnesses, questions, and interprets the world for consumption by another. Indeed, this is the very self-image many journalists are having to defend as they envisage how their profession will develop in an increasingly automated arena. But if those processes are absent, as in articles produced by AI, while the resulting output may be convincing, it is not clear to me that it can be considered journalism.

My question for discussion then is: If the processes of witnessing, interrogating, and attempting to understand an event are absent from AI-generated content, can the resulting output really be considered journalism?

Please share your thoughts and let me know what you think below.