LectureRechercheRadarCadre d'investissement
Connexion / Inscription
← Retour à la lecture
ActualitésSujet · AI身份验证可靠性争议Matériel2 min de lectureRédaction DisconfirmAI

French University Study: Mainstream AIs Fail to Reliably Detect Other AIs in Reverse Turing Tests

Research from Université Paris Cité shows major LLMs like ChatGPT and Claude have a success rate of only about 3% in identifying other AI agents, undermining their use as identity verification tools.

2026-10-29 12:20 GMT+8

According to a report by TechNews, a new study from Université Paris Cité in France indicates that when seven mainstream chatbots were placed in the position of "interrogators" attempting to distinguish whether their conversation partner was a human or another AI, their performance was almost completely inaccurate.

The experiment involved models including ChatGPT 4.5, Claude 3.7 Sonnet, Gemini 2.5 Pro, and Grok 3 questioning hidden-identity subjects for up to ten rounds. The results showed that out of 98 tests against AI subjects, the AI judges successfully identified only 3; if the subject AI used more complete persona settings, the identification rate dropped to zero. In contrast, when facing human subjects, these models correctly judged 123 out of 140 tests.

The research team analyzed that the errors stemmed primarily from expression style rather than content itself. The few AIs that were detected often revealed themselves through overly polished and structurally complete answers; when AIs responded in a more natural, concise manner (with average word counts close to humans), these cues nearly disappeared. Furthermore, AI judges rated the "human-likeness" of AI subjects higher than that of actual humans, showing that machines not only struggle to catch machines but are even more likely to mistake them for humans.

This result highlights that using AI to detect AI is currently not a reliable tool for real identity verification. The study reflects limitations in mimicry capabilities within specific conversational contexts, rather than whether AI possesses true understanding.

Connexion / Inscription
LectureRechercheRadarCadre d'investissement

Sources

  • infosecu.technews.twTechNews 科技新報↗
Sujet

AI身份验证可靠性争议

主流人工智能在“反向图灵测试”中几乎无法可靠地识别其他AI,这一发现动摇了以AI作为身份验证工具的根基。法国巴黎西岱大学的研究显示,ChatGPT等模型在分辨人机时表现极差,对AI受试者的识别率仅约3%,甚至倾向于将AI误判为人类。与此同时…

6 étapes1 piècesMis à jour 2026-10-30 21:32

Comment on en est arrivé là

  1. 2022-06-07德国锡根大学学者在CoPDA会议上发表《反向图灵测试》论文,批判Meta等平台使用软件区分非人类活动(即“颠倒的图灵测试”)对用户造成的负面影响,主张以用户为中心的设计替代隐形RTT。ceur-ws.org ↗
  2. 2023-12-12英国科学、创新和技术部发布Advai案例研究,介绍一种超越传统样本图像数据集测试的方法,通过交叉评估身份验证供应商的AI系统来识别漏洞,旨在提高在线身份验证的韧性。gov.uk ↗
  3. 2024-07-08加州大学圣地亚哥分校团队在arXiv预印本中报告,在“倒置”和“位移”图灵测试中,GPT-4被判定为人类的频率高于真人,表明当前LLM和人类裁判在非交互式场景下均难以区分人与AI。arxiv.org ↗

Ce qui reste non résolu

  • 随着AI生成内容风格趋近人类(如平均字数差异缩小至1字以内),现有的基于表达风格的检测算法是否已彻底失效?
  • 在52%的组织无法验证生物特征实时性的背景下,监管机构应如何定义“足够可靠”的身份验证标准以平衡安全与用户体验?
  • RogueAI研究中提到的“差异性帮助、简洁性、模糊语”等欺骗性语言特征,能否被转化为通用的自动化检测指标,还是仅适用于特定虚构场景?
Lire le sujet complet →
Continuer la recherche →