Luddites as gatekeepers of literature, By Osmund Agbo

The Luddites were a movement of skilled textile workers in England who rebelled against industrial machinery between 1811 and 1816. Having spent years mastering their crafts, they watched machines produce textiles faster and more cheaply, threatening their livelihoods and diminishing the value of skills painstakingly acquired over generations. Some, invoking the legendary General Ned Ludd, responded by smashing the machines. The British government retaliated with military force, executions and harsh punishments. Ultimately, mechanised production prevailed, and the traditional methods the Luddites sought to preserve gradually gave way.

Their story offers a useful lens through which to examine today’s anxieties about artificial intelligence and literary creation. As machines become increasingly capable of producing fluent, sophisticated prose, a familiar battle is unfolding between those who embrace technological change and those who fear it threatens the very foundations of their craft. The question is whether we are witnessing a legitimate defence of literary integrity or the emergence of a new class of Luddites determined to police how writers create.

Recently, there was a controversy surrounding Jerry Falade’s highly anticipated debut crime novel, Call Me, I’ll Hide the Body. The manuscript reportedly attracted an offer exceeding $2 million from Minotaur, a Macmillan imprint, in a 14-way auction, with British publishers also making substantial bids. Publication was planned for 2028. Yet the deal collapsed after concerns emerged about possible AI involvement in the writing. 

According to the author’s agent, the manuscript had initially impressed its readers, and the agency had declined to use AI-detection software because the author’s explanation appeared credible. Doubts reportedly intensified after a meeting in which aspects of that explanation appeared to change. The agency subsequently withdrew the manuscript, although the full circumstances remain unclear.

The case raises important questions about authorship, honesty and the limits of technological detection. It also illustrates how quickly an emerging literary career can become entangled in a dispute over the provenance of words.

Falade is not alone in facing controversy. Jason Arday, Mia Ballard and HM Wolfe, all black, have encountered scrutiny or backlash over allegations involving plagiarism, AI use or questions of authorship, although the circumstances differ considerably. Controversies involving  these Black writers and academics have unfolded amid a wider political climate in which anti-woke campaigns have sought to discredit individuals and institutions associated with progressive causes. These cases should not be conflated, nor should criticism of an author’s conduct automatically be dismissed as politically motivated. But neither should technological suspicion become a convenient instrument for discrediting writers without adequate evidence.

AI-detection platforms such as Pangram have entered this contested terrain, promising to distinguish human-written prose from machine-generated text. For sure, such tools may be useful in identifying passages that warrant further examination. But a detector does not have privileged access to an author’s mind or creative process. It identifies patterns associated with AI-generated language; it cannot, by that fact alone, establish plagiarism, dishonesty or literary worth. 

A human writer may produce prose that a detector mistakenly flags as artificial.

My concern, however, extends beyond the racial politics surrounding some of these controversies. The more fundamental question is how we should use a transformative technology without surrendering the values that make literature worth reading.

We need to stop treating AI as some evil spirit destined to possess and destroy the creative mind. The more useful question is not whether writers should use AI, but to what extent, under what conditions and with what obligations of disclosure. We need rules that distinguish legitimate assistance from wholesale substitution, intellectual collaboration from misrepresentation, and editing from the outsourcing of authorship.

Demanding that writers reject AI altogether is like asking a man to prove his manhood by walking 200 miles through the unforgiving desert sun when an aeroplane can take him to the same destination in less than an hour. The analogy is imperfect, because literary creation is not simply a journey from one point to another. The struggle, the choices and the craftsmanship may be integral to the destination. Nevertheless, insisting that every task must be performed the old-fashioned way merely because it once was is not a convincing defence of artistic excellence.

A writer may develop an original argument, conduct the research, determine the structure, supply the examples, and decide what the work should communicate. The writer then uses AI to improve sentence construction, eliminate repetition, or clarify a difficult passage. Has the writer committed a sin? Not inherently. Has the writer surrendered literary integrity? Not necessarily. Many of us still run our writing through Microsoft Word to identify spelling and grammatical errors.

We must also distinguish AI assistance from plagiarism. Plagiarism involves presenting another person’s words, ideas or creative work as one’s own without appropriate acknowledgment. AI-generated text does not automatically constitute plagiarism, although AI can reproduce material from existing works without attribution. That is a genuine concern, but it is not the same as declaring every use of generative AI an offence against literary creation.

Two writers described childhood poverty in a Nigerian village. One grew up hungry, watched classmates arrive at school with money while he had none, and learned early that poverty determines not only what a child owns but also how the world treats him. Years later, he transforms those memories into a novel. The other has never experienced such deprivation but uses AI to produce an eloquent, emotionally compelling account of the same experience.

Both passages might be beautiful. Both might move readers. Yet their literary authority may differ. The first draws upon lived experience, memory and personal reflection; the second may depend more heavily on the writer’s ability to direct and refine a machine’s output.

But we should not romanticise suffering or human authorship. Having endured hardship does not automatically make someone a great writer, and a novelist need not personally experience every circumstance described in a book. Imagination has always allowed writers to inhabit lives beyond their own. What matters is the intelligence, insight, judgment and artistic choices brought to the work.

Photography did not abolish painting, nor did the arrival of digital cameras make every photograph a work of art. Technology changed how art could be produced without eliminating the need for artistic judgment. AI may similarly transform literary production without making originality irrelevant.

The danger is that, in the absence of agreed standards, we have begun using AI as both the alleged offender and the judge. There is no universally accepted framework governing how much AI assistance a literary work may receive, when disclosure is necessary or how accusations should be investigated. Publishers, universities and literary competitions need transparent policies proportionate to their purposes. A competition may reasonably demand unaided writing, while a publisher may permit AI-assisted editing provided the author remains responsible for the work and complies with disclosure requirements.

The Luddites might have achieved more by demanding protections against exploitation and a fairer distribution of the gains from industrialisation than by trying to destroy machinery they could not ultimately stop. Their struggle was not simply against technology; it was against the economic power technology placed in the hands of those who controlled it.

Literature faces a similar choice. We can develop sensible rules that protect authorship, reward originality and guard against deception, or we can elevate detection software into an unquestionable authority and allow suspicion to replace judgment.

The purpose of literary gatekeeping should be to protect the integrity of literature, not to preserve particular methods of producing it. Our task is not to keep machines out of the writer’s room. It is to ensure that the human mind remains in charge of what comes out of it.

Osmund Agbo is a medical doctor, author, and the 2026 recipient of the Thabo Mbeki Prize for Scholarship and Public Service