ACADEMIC SLOP The Human Integrity Crisis Misattributed to Artificial Intelligence-PNBA Identity Physics Substrate-Neutral Structural Foundation Laws

The phrase "AI slop" has entered academic discourse as shorthand for low-quality research output. This framing is structurally incorrect and, for certain populations, constitutes a civil rights violation. The retraction crisis predates generative AI by decades. Its causes — paper mills, compromised peer review, citation laundering, and institutional gatekeeping — are executed by credentialed humans using institutional email addresses, not by AI systems operating autonomously. A May 2026 Nature-published audit of 2.5 million biomedical papers and 97 million references found the fabricated citation rate in 2025 was more than twelve times greater than in 2023 — with fabricated references appearing in papers that passed human peer review. At NeurIPS 2025, the world's most prestigious AI conference, 100 hallucinated citations passed review by 3 to 5 expert humans per paper. The human reviewers missed them. An AI scanner found them. The failure of due diligence belongs to the humans who reviewed, approved, and published the work now being retracted. This paper presents that evidence, names the structural failure accurately as Academic Slop, and argues that restricting AI use for communication and translation purposes — particularly for neurodivergent and disabled researchers — constitutes discrimination under the Americans with Disabilities Act, Section 504 of the Rehabilitation Act, and Executive Order 14303. The solution is not to restrict the tools that help marginalized researchers participate. The solution is to fix the peer review system that let the slop through. 1. The Numbers Do Not Lie — And They Predate ChatGPT The retraction crisis is real. The scale is documented. What is not documented anywhere in the mainstream framing is that it began long before generative AI existed. The Retraction Watch database contains over 61,000 retraction records as of early 2025. In 2023 alone, over 14,000 retraction notices were issued — a single-year record. In 2024, more than 9,000 additional retractions followed. The median time from publication to retraction is 562 days. That interval is not a data anomaly. It is a measurement of how long peer-reviewed, editor-approved, institutionally credentialed work sits in the literature contaminating subsequent research before anyone catches the problem. The citation contamination problem is now precisely quantified. A study published in The Lancet in May 2026 and reported in Nature audited 2.5 million biomedical papers and examined 97 million references. It found nearly 3,000 papers containing fake citations — references to publications that do not exist. The rate of fabricated citations in 2025 was more than twelve times greater than in 2023. A separate Nature analysis estimated that approximately 1.6% of all publications from 2025 contained at least one non-existent reference. The authors described their findings as "conservative underestimates" of the true scale. Every one of those papers passed peer review. Every one was submitted by a human with an institutional affiliation. Every one was approved by a human editor. The AI did not submit the paper. The AI did not approve the paper. The AI did not assign it a DOI and index it in PubMed. The peer reviewers did that. 2. The NeurIPS Test Case: Peer Review Fails Where AI Detection Succeeds The NeurIPS 2025 incident is the cleanest available test of where the integrity failure actually lives. NeurIPS — the Conference on Neural Information Processing Systems — is the world's flagship machine learning conference. In