Joe X Young

10/06/2026

"Dumpster Fire" author lists are making the rounds again.

Recently I’ve seen the promoting of a list identifying people who are considered “Dumpster Fires” within the community. The list is intended as a warning to emerging authors. Yet such a list raises certain questions. Our community may have good reason to want to protect its members from harm, but it also has responsibilities to fairness, accuracy, proportionality, and the issue of whether a person on the list has provided any indication of making amends for their previous actions.

Supporters of these lists usually begin from a practical reality: as there is no official governing body and no rules of conduct for the horror writing community it is not truly capable of protecting people. In many cases, those who experience harassment, coercion, manipulation, retaliation, or repeated personal boundary violations find that they are on their own. Conferences, publishers, workshops, social media groups, and organizations may lack clear procedures, may be reluctant to intervene, or may prioritize reputation management over member safety.

A list can help people see patterns that would otherwise remain hidden. One allegation or uncomfortable interaction may be dismissed as a misunderstanding; a pattern of similar reports can reveal that something more serious is happening. Unfortunately these lists can collapse very different things into one category as “Dumpster Fire” can mean many things: abusive behaviour, professional misconduct, ideological disagreement, rudeness, manipulation, physical boundary violations, public controversy, or simply being disliked.

When a community does not clearly define what qualifies someone for inclusion, a list can blur the distinction between severe harm and ordinary conflict, and it can leave readers to infer more than the evidence supports. The list I’ve seen doing the rounds over the past few days is such an example, as I know the back-stories of the majority of the names on the list, some of whom have made a single significant mistake, whilst others are serial offenders with proven track records of extreme and deliberately abusive behaviour.

There is also the problem of verification as lists can spread faster than the context needed to interpret them. An allegation may be true, partly true, mistaken, exaggerated, or impossible for outsiders to assess. Once names circulate, however, reputational damage can become irreversible, even if later information complicates the story or even exonerates the person named.

That raises the following questions around governance:

Who gets to maintain the list? Although I have seen calls for something similar to ‘Writer Beware’ but aimed at author misconduct several times over the past few years, as far as I am aware there is no single ‘go-to’ source for one. In my opinion it’s not enough to warn a community that a person or entity is ‘bad’, one must provide a valid reason with evidence for saying so. We have ‘Writer Beware’ serving the community with warnings against the nefarious activities of certain publishers, agents and editors. It takes a lot to maintain and fact-check. Having a ‘Community Beware’ service could rapidly get out of hand with people reporting others for the slightest of personal issues. It could eventually be the case that everyone in the community would be reported by at least one other person, even me, and I’m really nice… yeah, okay, maybe not, but I think you get the point.

What evidence standard would such a list use? The list currently promoted doesn’t provide evidence, it doesn’t even state a specific issue, it just names people and says that if you want to know why they are on the list then it’s up to you to look the people up, which could prove fruitless via restricted accounts, deleted posts etc. the list posts often have people rushing to check their friend lists and removing people without further scrutiny simply because their name is on a list in spite of no actual evidence of wrongdoing being presented.

Can someone challenge an entry? Of course, however it’s often the case that once you’ve made the list, even if you’ve made an effort to turn your life around since, you’re basically cooked. Those acting as ‘Judge, Jury, and Executioner’ are not willing to reassess based on a redemption arc, they just want people blocked, banned, or otherwise ostracized. No extenuating circumstances, no second chances, no exceptions.

Is there a difference between a private safety warning shared with care and a public blacklist with no oversight? I believe there is, but it’s marginal, and both have their problems. Accountability systems require structure rather than ad hoc decision-making. Informal lists concentrate power over someone’s fate in a few hands while appearing to speak for a whole community. Such lists can be challenged as those on it would similarly argue that those promoting it are the problem, so without presenting absolute facts with proper documented evidence it’s one person’s (or group’s) word against another.

Part of what makes this debate so difficult is that our community is trying to solve more than one problem at once. We want to protect the vulnerable. We’d prefer people to report harm but also want fair standards to avoid Kangaroo Courts and lynch mobs as accusations can carry life-changing consequences.

If the goal is accountability, the community may need clearer codes of conduct, reporting channels, and response procedures, which although such things exist in certain groups, organizations, and personal websites, the open community in its current form doesn’t have, doesn’t appear to want, and wouldn’t be legally enforceable.
Lists of “Dumpster Fire” people emerge because communities are trying to navigate real danger, real mistrust, and real failures of accountability. People need ways to protect one another and recognise patterns of harm before they escalate, but any serious discussion should distinguish between at least four categories: allegations of serious harm, patterns of manipulative or coercive conduct, professional unreliability, and ordinary personal conflict. A community that labels all four simply as “Dumpster Fires” risks losing clarity as a list that is vague, poorly governed, or impossible to challenge may reproduce exactly the abuses it hopes to prevent.

30/03/2026

Fact-Checking the Cancellation of ‘Shy Girl’ by Mia Ballard

For those of you who haven’t heard about the case or wish to avoid paywalled news, in March 2026, Hachette Book Group cancelled the U.S. publication of the horror novel ‘Shy Girl’ by Mia Ballard and discontinued its U.K. edition following widespread online allegations that generative AI had been used heavily in the book’s creation. This marked what major outlets described as the first known instance of a Big Five publisher pulling a commercial novel over suspected AI involvement.

The novel was originally self-published by Ballard in February 2025. It gained initial popularity, accumulating approximately 4,900 Goodreads ratings with an average of 3.52 stars. Hachette acquired North American rights in July 2025. The U.K. edition (Wildfire imprint) was released in November 2025 and sold roughly 1,800 print copies according to NielsenIQ BookData. A U.S. release had been scheduled for April 2026.

Online speculation about AI involvement began months earlier on platforms including Goodreads, Reddit, and YouTube. Reviewers and readers pointed to “nonsensical metaphors,” “odd, repetitive phrasing,” “bizarre formatting,” “typos,” and prose that “resembled output from AI tools” or sounded like “ChatGPT.” A January 2026 YouTube video essay titled “I’m pretty sure this book is AI slop” (nearly three hours long) analysed the text line by line and amassed over 1.2 million views.

AI detection tools provided quantitative claims. Thad McIlroy did not originate the initial suspicions (those came from readers, a YouTube essay, and Reddit threads months earlier). He amplified them by independently testing the text with multiple AI detectors (including the Pangram result cited in coverage) and directly supplying the story and evidence to The New York Times. His role turned private/online speculation into a high-profile scandal that prompted Hachette’s swift cancellation.

This places him as a key behind-the-scenes figure rather than a neutral observer. No evidence suggests he had any direct editorial, publishing, or personal connection to Ballard or Hachette prior to this. He has positioned himself as both a catalyst for accountability on AI in publishing and a commentator critical of how the publisher handled the author. In his own blog he wrote "Hachette threw Mia Ballard under the bus, sullied and cancelled without a chance to defend herself in the court of public opinion."

The New York Times analysed selected passages using multiple AI detection tools and identified recurring patterns it described as characteristic of AI-generated text, including gaps in logic, excessive use of melodramatic adjectives, and overreliance on the “rule of three” in writing.

No conclusive evidence, such as direct admissions or verifiable source files, has been presented to prove the involvement of A.I. in the creative process.

Experts in digital forensics and literary analysis have noted that features attributed to A.I. could equally be the result of stylistic choices, editing mishaps, or artistic experimentation by the author and designer. In tests pertaining to student papers, A.I. has been shown to exhibit an accusation bias against certain groups:

Non-native English speakers (EAL/ESL writers): Studies, including from Stanford, show detectors are far more likely to flag their work as AI-generated due to differences in phrasing, formality, or learned structures. One analysis found over 61% of TOEFL (test of English as a Foreign Language) essays misclassified, with 97% flagged by at least one tool.

Neurodiverse writers and those using certain dialects or stylistic patterns: Their writing may appear "formulaic" or repetitive to detectors trained primarily on standard native-speaker examples.

Other demographic disparities: Some reports note higher accusation rates for Black students or other underrepresented groups, potentially exacerbating existing educational inequities.

Hachette stated that it conducted a “thorough and lengthy” internal review of the text in the weeks leading up to the decision. On March 19, 2026, one day after The New York Times approached the publisher with evidence of apparent AI generation, Hachette pulled the book. The title was removed from Amazon and the Hachette website that same afternoon. A spokeswoman said: “Hachette remains committed to protecting original creative expression and storytelling.” The company’s policy requires submissions to be original to the authors and asks authors to disclose any use of AI during the writing process.

Author’s Response:

In an email to The New York Times sent late on March 19, 2026, Mia Ballard denied using AI to write ‘Shy Girl’. She stated that unbeknownst to her, an acquaintance she hired to edit the self-published version had used generative AI, and “changed a lot of the wording.” Ballard added that she did not perform a final review pass due to time constraints. She wrote: “This controversy has changed my life in many ways and my mental health is at an all-time low and my name is ruined for something I didn’t even personally do.” She noted she was pursuing legal action and could not elaborate further on the editing process. She appears to have since taken down her website and all social media accounts.

No public evidence has emerged of Ballard admitting to ‘direct’ use of AI for writing the core manuscript. Some accounts reference her earlier social media or interview responses denying A.I. allegations outright.

Allegations regarding the cover art:

The self-published edition’s cover featured a cropped and edited portion of the painting ‘Dreamer’ by Scottish artist Whyn Lewis, sourced from Pinterest. Ballard did not initially credit Lewis or obtain permission. The artist contacted Ballard after the book gained attention; accounts differ on the exact resolution, with some indicating compensation or removal efforts, but the issue resurfaced publicly. Hachette’s U.K. and planned U.S. editions used new cover designs. These new covers were reported in some online comments speculating that they were “heavily inspired” by Lewis’s work.

So, those are the facts. Now on to my opinion.

I have seen other posts which are claiming this entire situation is a ‘witch hunt’ started by an ‘old white guy’ targeting a young black woman. I’m not saying racism isn’t a thing, I’m saying racism isn’t the core issue here, nor should it be, as anyone using generative A.I. for novel content deserves whatever negativity they get, regardless of ethnicity. As well as being factually incorrect, I think blaming Thad McIlroy and presenting this as a race issue is divisive ‘low hanging fruit’ that is solely aimed at stirring resentment to score comments and likes.

This is not to say that McIlroy isn’t in part responsible for the cancellation of ‘Shy Girl’ as he did massively escalate the issue, but the evidence clearly shows that the accusations against Mia Ballard do have a basis in fact and began months before McIlroy got involved. Thad McIlroy’s background shows that he has worked in and around publishing for over 30 years and has consistently reported on the use of A.I. in publishing having written over 500 articles and a dozen books on the subject. Which leads me to suspect that his analysis of Mia Ballards work was a part of his regular line of work and was in no way personal or indeed racist.

There is a question of legality when it comes to those who have fed her work into multiple instances of A.I. to achieve their conclusions, for which permission must surely have been required. Perhaps Ballard has some recourse via that route against all who submitted her manuscript into A.I. checkers. There’s also the person who read Ballard’s entire manuscript line by line on YouTube for commentary, that’s potentially a breach that falls outside of the usual scope of ‘fair use’ doctrines.

Mia Ballard is not denying that generative A.I. was used to create portions of the text of ‘Shy Girl’, her claim is that an as yet unnamed ‘editor’ used generative A.I. and Mia didn’t notice it. Had she re-read her manuscript instead of being in a rush to get it on shelves, she may well have caught the issues and confronted the editor over them. Her denial isn’t that the book wasn’t, even in part, generated by A.I., her denial is that she didn’t personally do it. So, taking her own admission that the book does have generative A.I. into account then all of the accusations do seem to be based in the facts as we already know them.

Do I believe that Mia Ballard deliberately used A.I. to generate her book?

I’m on the fence there, as so far I’ve not seen any accountability from the editor she trusted to edit her manuscript. This to me is a crucial piece of the puzzle as the editor should at the very least be brought forward to explain what was done and why. Until an actual human being comes forward, we have no factual basis for Ballard’s accusations regarding her editor, or even that the editor actually exists as a human being. Even if the editor does come forward to provide clarification, then it’s my opinion that it should still have been Ballard’s responsibility to re-read the manuscript, not only to accept the changes, but also to ensure that her ‘voice’ had not been lost in the edits. Writers are, after all, ultimately responsible for making sure that their published book is actually theirs.

Do I believe that Mia Ballard can successfully sue the NYT, Thad McIlroy, Hachette etc?

That’s a ‘mixed bag’ as it really depends on what she intends suing for. If it’s defamation then that would be a ‘no’ as A.I. was, by her own admission, used to generate parts of her book, so she can hardly call out people for using facts.

If she sues Hachette for breach of contract, then that depends on the wording of the contract. If their general policy for written works from authors is ‘no A.I.’ then she will have a struggle on her hands with that one too. Hachette may well have been lax in their treatment regarding ‘Shy Girl’ but as generative A.I. was, by Ballard’s own admission, used in part, then in my logic-based opinion (I am not a legal professional) Hachette has a strong defence position as copyright law clearly dictates that all ‘creative content’ must be of human authorship otherwise it cannot gain a full copyright. In other words, Hachette cannot realistically be expected to release ‘Shy Girl’ with FULL copyright ownership as neither Mia Ballard or Hachette are responsible for those portions of text generated by A.I..

If Mia Ballard wishes to sue others for defamation, then she would also have to provide evidence that individual accusations were false, so she may get somewhere with some of those as not everyone sticks to the facts.

As for the theft of art for the cover, it’s common for authors to steal art, or indeed pieces of it, for use on their books. I’ve seen many examples of it across many genres, and there are no excuses for it as even the most routine of searches would have shown that it’s illegal and unethical. Mia Ballard’s use of someone else’s art without permission isn’t a good look, but in my opinion, it also should not be ‘career-ending’. Similarly, if the reality is that Mia Ballard is guilt of nothing more than negligence in generative A.I. being used in the text of her book, then she should be given the opportunity to return her manuscript to its pre-edit state and hire a better editor. I believe that even though this is high-profile within writerly circles the damage to her career from reader perspectives isn’t so severe that she couldn’t have survived this given that the majority of the reading public would be unaware of anything being amiss.

People are finding out the hard way that poorly written books and A.I. generated slop WILL end careers, rightly so, but we need to proceed with caution. The ‘Court of Public Opinion’ seems to have transitioned into ‘The Kangaroo Court of Judge, Jury, and Executioner’, showing no mercy and heading straight to a ‘Guilty’ verdict without seeking the truth behind the headlines.

I live for the time when we are all working together as a better community, when we become aware that we’re not the only ones who are allowed to make mistakes, and as such use our knowledge and guidance to help others to learn from theirs.

Mia Ballard made mistakes. They have cost her dearly.

It's a lesson for all.

27/02/2026

CC Adams invited me to talk about what I do and why I do it, so if you want to find out why he said: "If ever there were an individual to be described as a genre visionary, Joe 'X' Young would be in the running." then check out the link below!

If ever there were an individual to be described as a genre visionary, Joe 'X' Young would be in the running. For members of both the Get Writing Horror and Get Reading Horror, his passion and commitment are undeniable. Always with an eye on the big picture for horror writers to gain both traction a...