In today's rapidly evolving digital landscape, AI language models like ChatGPT have democratized access to knowledge, streamlined professional workflows, enhanced creative expression, and delivered personalized assistance at an unprecedented scale. By providing real-time feedback, stylistic refinement, and structural guidance, these tools have revolutionized the writing process, empowering both seasoned writers and newcomers to communicate with greater clarity, efficiency, and nuance. Yet despite these remarkable capabilities, AI systems can still "hallucinate"—generating confident but false information—highlighting the ongoing need for careful human oversight and verification.
The proliferation of AI-powered writing tools has sparked intense debate about academic integrity and intellectual authenticity. Critics worry that unrestricted AI use could erode original thought and independent learning. This concern has led to the development of AI detection software designed to distinguish human-generated text from machine output. However, this has created an arms race: in response, a new wave of "humanizing" tools has emerged, specifically designed to bypass these detectors by making AI-generated text appear more traditionally human. While both detection and humanizing tools ostensibly aim to preserve authenticity, they have created unexpected and serious challenges for non-mainstream writers.
These challenges particularly affect two significant groups: neurodiverse individuals and those for whom English is a second language. Both communities now face a higher risk of having their authentic writing incorrectly "flagged" as computer-generated. The root of this problem lies in how AI detectors operate: they typically rely on metrics like "perplexity" to assess writing fluency and complexity. However, these systems were trained primarily on neurotypical, native English writing samples, creating inherent biases against alternative writing styles and patterns.
The scope of this bias is striking. Research from Stanford University revealed that AI detectors misclassified over half of TOEFL essays (written by non-native English speakers) as AI-generated, with some detection systems flagging as much as 97% of these essays. Such misclassifications do more than undermine the hard work of non-native speakers—they can devastate academic careers and professional opportunities, creating barriers to advancement based on false premises.
The impact on neurodiverse writers is equally concerning. Many neurodiverse individuals express themselves in ways that diverge from mainstream conventions. For instance, writers with autism may prefer direct, literal language or maintain intense focus on specific topics. While these writing styles are entirely coherent and meaningful, AI detection tools often flag them as "robotic" or "unnatural," leading to false accusations of AI generation. In academic and workplace settings, these erroneous flaggings can have serious consequences, from accusations of academic dishonesty to the dismissal of valuable contributions.
Paradoxically, while AI detection tools can create barriers, other AI technologies have become invaluable assistants for many neurodiverse writers, helping them structure thoughts, refine grammar, and clarify their meaning. Writer and accessibility advocate Taylor Arndt, who is both neurodivergent and blind, offers a compelling illustration of how AI can enhance the writing process through a thoughtful, structured approach (*link to her article below).
For Arndt, AI functions as more than a tool; it's a collaborative partner. Her process reveals the thoughtful way AI can enhance rather than replace human creativity. She begins with a raw, unfiltered draft to capture her core message and key points—a stage where her unique voice and perspective shine through unedited. Then she uses AI to refine grammar, reorganize ideas, and suggest phrasing improvements. "It's like having a personal editor who helps me shine without changing my voice. The AI doesn't replace me—it enhances my ability to communicate." After AI assistance, Taylor carefully revisits each draft to ensure it maintains her authentic voice: "AI helps with structure and clarity, but it's still my heart and soul on the page. The final product needs to feel like me."
This three-stage process—raw drafting, AI refinement, and personal editing—mirrors long-established practices in academic and professional writing. Consider how a professor develops a paper for publication: they begin with initial concepts and outlines, delegate portions to graduate researchers for literature reviews and data collection, receive drafts of preliminary sections from research assistants, synthesize these contributions, seek peer review feedback, engage professional copyeditors, and make final revisions. In both cases, the final product emerges from a collaborative process that enhances rather than diminishes the authentic scholarly contribution.
The parallel between AI assistance and traditional academic support structures highlights a crucial point: the myth of solitary creation has always been just that—a myth. Whether leveraging AI tools or human collaborators, the path to polished, effective communication typically involves multiple contributors and refinement stages. The key difference is that AI provides immediate, accessible assistance compared to traditional human collaborators, democratizing access to the kind of editorial support that was once available only in professional or academic contexts.
This reality creates a troubling paradox in our current landscape. When writers feel concerned about submitting work developed with AI assistance, they might turn to AI-driven "humanizing" services designed to evade detection. Ironically, this process often results in text that is more thoroughly machine-processed but more likely to evade AI detection algorithms, while the original human-AI collaboration—though more authentic—remains susceptible to being flagged as AI-produced. This situation not only erases the writer's authentic voice but reinforces the flawed standards of detection software.
All this raises a fundamental question: what constitutes "authentic" writing, and who has the authority to decide? These issues connect to deeply rooted cultural biases about creativity. Western thought, influenced by Romantic ideals, has long romanticized the solitary genius who produces wholly original work. Yet historical evidence consistently shows that genuine innovation has always been collaborative, whether through Shakespeare borrowing plots, musicians building on existing melodies, or scientists advancing their predecessors' research. AI's role simply highlights this eternal truth: creativity doesn't happen in isolation and never has. Rather than viewing AI as a threat to authenticity, we should recognize it as a powerful extension of human capability—another step in humanity's long tradition of leveraging tools to enhance intellectual and artistic output.
The disconnect between idealized originality and actual practice becomes particularly apparent in academic settings. At the undergraduate level, institutions often emphasize strict individual originality, expecting each student to produce entirely independent work. Yet in professional academic contexts, the reality is far more nuanced. Professors regularly delegate literature reviews, data collection, or preliminary draft sections to student researchers, then synthesize and refine these contributions into final documents. While students might receive acknowledgment for their efforts, they don't necessarily gain co-authorship, reflecting established hierarchical norms. Similar patterns exist in legal contexts, where judges rely on clerks to draft significant portions of opinions that ultimately bear the judge's name alone. These practices demonstrate how "originality" often takes more collective forms in professional settings.
Moving forward requires a fundamental shift in how we teach, assess, and reward creativity. AI is not merely a convenience—it's a collaborative partner that amplifies human potential. Instead of fixating on whether work is "original" in the traditional sense, we should evaluate whether the final product represents a meaningful synthesis of human insight and AI assistance. Rather than penalizing students for using AI, we must teach them to use it responsibly, preparing them for a future where human-AI collaboration is the norm.
The intersection of AI and neurodiversity perfectly illustrates both the promise and pitfalls of this new era. While current AI detection tools may disadvantage non-native and neurodiverse writers, AI itself can empower these same individuals to express their ideas more effectively. The growing reliance on additional AI tools to "humanize" already human-driven work underscores the urgent need for more inclusive, equitable standards of assessment. By shedding outdated notions of originality and embracing AI as a creative partner, we can design a future that celebrates diverse voices, fosters innovative collaborations, and redefines authenticity for the digital age.
Ideally, the future of creativity is not a competition between humans and machines, it's a partnership that allows both to thrive – or even more, a tool for giving voice to, and amplifies, human creativity that has heretofore been unexpressed or muted. This kind of partnership is not new but actually mirrors established collaborative practices in professional and academic contexts while it serves to democratize access to sophisticated editorial support. By recognizing these parallels and addressing current challenges while embracing the opportunities that AI presents, we can create a more inclusive and dynamic creative landscape that benefits all participants, regardless of social station, cognitive styles, or linguistic background. As Arndt writes, "for people like me, with their particular set of challenges, AI is more than just a tool; it’s a necessity.
This means we should look at AI altogether differently instead of threatening to punish or stigmatize its use. The technological fact of AI is here to stay, and its potential for empowering people is huge. For me, it’s a bridge between my ideas and the outside world.
__________
*Read Taylor Arndt’s article at medium.com
Disclosure
This article was authored by JR Grimes, an autistic scholar and founder of The Ibesian Magazine, through a rigorous academic writing process. Grimes conceived the article's core arguments and key insights, utilizing AI platforms (such as ChatGPT) in roles analogous to research assistants and professional editors. The development involved multiple iterative drafting cycles, with AI providing targeted assistance and suggestions related to structural organization, clarity enhancement, and stylistic refinement. While AI contributed to improving the manuscript's technical and literary qualities, all substantive arguments, analytical frameworks, and scholarly conclusions represent Grimes's original intellectual contribution.
This collaborative approach reflects standard academic practice, whereby scholarly authors routinely work with various contributors to distill and strengthen their research findings and presentation. By explicitly acknowledging AI's role in this work, the author seeks to advance transparency while exemplifying productive and ethical models of human-AI collaboration.
Copyright (2025) is asserted in accordance with the January 2025 U.S. Copyright Office report, Copyright and Artificial Intelligence, Part 2: Copyrightability, which affirms that this work qualifies for copyright protection as an original creative product of the author, with AI functioning solely as an assistive tool.