Dr Broadly, wearing my scholarly hat, your findings deeply concerned me. The retroactive classification of a 12-month-old piece as 99.8% AI-generated is not just an error. I see this as an epistemological crisis. Such a shocking occurrence demands an urgent interrogation of the ontological and methodological foundations upon which these AI detection systems operate. If authorship can be arbitrarily redefined by opaque and inconsistent tech algorithms, then the very nature of intellectual labor is under siege. This is not simply a technical flaw but a profound misalignment between human creativity and mechanized gatekeeping. Left unchecked, these flawed metrics risk eroding the legitimacy of human expression, rendering the distinction between human and machine-generated content a matter of shifting, algorithmic decree rather than objective evaluation. As I pointed out in 2021 I have no fear of artificial intelligence but if these systems try to be artificial superintelligence to control humans, it gives me nightmares. https://medium.com/illumination/artificial-intelligence-does-not-concern-me-but-artificial-super-intelligence-frightens-me-e5354737d119

The implications of this nonsense extend beyond individual injustice. They threaten the fabric of knowledge production and the ethical stewardship of digital publishing. This madness must not merely be stopped but it must be rigorously deconstructed before it redefines the future of authorship itself. I invite all scholarly writers to take a stand on this issue. I will lead this via an academic paper soon to illuminate society. Thank you for sharing this concernng case study. I loved your story about cancer cells covering ketosis which is a topic close to my heart. Copyleaks has no idea how a scientist distills half a century of knowledge for the public. Their technicians have no such capability. It is absurd to say a story 99.8 AI written while it is sounding like a real human. I am disgusted and disappointed with this outcome.