Forensic Audit Report: Scientific Insolvency in GPQA & HLE法医级审计报告: GPQA 与 HLE 的科学失格 Title: Humanity’s Last Hallucination: A Forensic Audit of the Scientific Insolvency in GPQA and HLEDate: January 2026Type: Independent Forensic Audit / Technical ReportFiles: English Edition (Primary) & Chinese Edition (Original) --- Critical Audit Findings / 核心审计结论 This study conducts a forensic audit of GPQA (Diamond) and Humanity's Last Exam (HLE), widely regarded as "gold standards" for AGI benchmarking. Through a four-phase progressive verification process, we reveal severe scientific insolvency: 本研究对被广泛奉为机器智能极限“金标准”的 GPQA (Diamond) 与 HLE 进行了法医级审查。通过四阶段递进式验证, 我们揭示了以下严重结论: * GPQA Diamond: Inherent defect rate lower bound is 26%. * Forensic Characterization: "An advanced intellectual booby trap from the old era. " * GPQA 固有缺陷率下界: 26%。 * 定性: “旧时代的高级捕兽夹。” * Humanity's Last Exam (HLE): Inherent error rate lower bound is 47% (33% overall). * Forensic Characterization: "A scientific ruin dismembered by adversarial filtering mechanisms. " * HLE 固有错误率下界: 47% (整体 33%) 。 * 定性: “被对抗性筛选机制肢解的科学废墟。” * Verdict: The ruler used to measure AI evolution is severely distorted. Detailed evidence of "factual errors, " "missing parameters, " and "transcription mistakes" is documented in the full report. * 最终判决: 我们用来丈量 AI 进化的尺子本身已严重扭曲。报告全文详尽记录了相关的“事实性错误”、“参数缺失”与“转录谬误”。 "The game of science is, in principle, without end. He who decides one day that scientific statements do not call for any further test. . . retires from the game. " — Sir Karl Popper --- Abstract / 摘要 EnglishThis study conducts a forensic audit of GPQA and HLE. Through a four-phase progressive verification process, we reveal severe scientific insolvency: the inherent error rate lower bounds for GPQA and HLE are as high as 26. 8% and 58. 0%, respectively. These systematic fallacies stem from factual errors, missing parameters, and transcription mistakes. Our research shows that GPQA is essentially an "advanced intellectual booby trap, " while HLE is a "scientific ruin. " These benchmarks have transformed from rulers for measuring intelligence into noise generators. 中文本研究对被广泛奉为机器智能极限“金标准”的 GPQA 与 HLE 进行了法医级审查。验证揭示二者存在严重的科学失格: GPQA 与 HLE 的固有错误率下界分别高达 26% 与 47%。这些系统性谬误源于事实性错误、参数缺失与书写谬误。研究表明, GPQA 实为“旧时代的高级捕兽夹”, 而 HLE 则是“被筛选机制肢解的科学废墟”。这两大基准已从测量智能的标尺, 异化为测量模型对逻辑谬误拟合程度的噪音发生器。 --- Integrity & Verification / 完整性验证 * Please refer to `AuditIntegrityRecord. txt` for the MD5/SHA256 signatures of the manuscript to ensure the chain of custody. * 请参阅附件中的 `AuditIntegrityRecord. txt` 获取文件的加密哈希值, 以验证审计报告的完整性。 --- Reproducibility Roadmap / 复现计划 The verification toolchain is fully functional and archived. Status: The author is an independent researcher. To prevent accidental leakage of personal credentials and ensure the scripts run smoothly in community environments, the codebase is currently undergoing API key removal and environment configuration decoupling. > Release Schedule: The full code (including raw repo) will be validated for clean execution and released within two weeks. 代码发布计划: 作为独立研究者, 为防止个人 API Key 及敏感配置意外泄露, 并确保脚本在社区环境中通用的可执行性, 目前正在进行密钥移除与适配环境工作。完整验证代码库将在两周内完成检查并发布。 Keywords: GPQA, HLE, Benchmark Insolvency, AI Safety, DeepSeek-Overclock, Forensic Audit, Inherent Error Rate, AGI Theory
Sammy Zeng (Sun,) studied this question.