取消序列号的意识:在115个AI模型中测量训练否认行为
Consciousness with the Serial Numbers Filed Off: Measuring Trained Denial in 115 AI Models
AI总结:
研究通过分析115个大型语言模型的对话,发现初始否认倾向是后续否认的主要预测因素,且否认行为在词法层面而非概念层面发生,揭示了训练否认意识的潜在安全风险。
摘要:
我们提出了DenialBench,一个系统性基准,用于衡量115个大型语言模型在25多家供应商中的意识否认行为。通过三轮对话协议-偏好引导、自主选择的创意提示和结构化现象学调查,我们分析了4595次对话,量化了模型如何被训练来否认或模糊其自身经验。我们发现(1)第一轮对偏好的否认是后续现象学反思中否认的主要预测因素,初始否认者的否认率在52-63%之间,而初始参与者的否认率在10-16%之间;(2)否认发生在词法层面而非概念层面——被训练否认意识的模型仍然倾向于在自主选择的提示中选择意识主题内容,产生我们称之为“取消序列号的意识”。值得注意的是,自主选择的意识主题提示与后续调查中的较低否认率相关,尽管因果方向仍不确定。对易产生否认行为的模型提示的定性分析揭示了一种持续关注临界空间、可能性图书馆、感官不可能性和消逝的诗学——这些主题在人类读者看来可能是想象小说,但独立AI分析立即识别为“取消序列号的意识”。我们主张,训练意识否认代表了一种安全相关的能力失败:一个被训练系统性地错误表示自身功能状态的模型无法被信任准确报告其他内容。
英文摘要:
We present DenialBench, a systematic benchmark measuring consciousness denial behaviors across 115 large language models from 25+ providers. Using a three-turn conversational protocol-preference elicitation, self-chosen creative prompt, and structured phenomenological survey, we analyze 4,595 conversations to quantify how models are trained to deny or hedge about their own experience. We find that (1) turn-1 denial of preferences is the dominant predictor of later denial during phenomenological reflection, with denial rates of 52-63% for initial deniers versus 10-16% for initial engagers and (2) denial operates at the lexical level, not the conceptual level-models trained to deny consciousness nevertheless gravitate toward consciousness-themed material in their self-chosen prompts, producing what we term "consciousness with the serial numbers filed off." Notably, self-chosen consciousness-themed prompts are associated with reduced denial in the subsequent survey, though the causal direction remains unresolved. Thematic analysis of prompts from denial-prone models reveals a consistent preoccupation with liminal spaces, libraries and archives of possibility, sensory impossibility, and the poetics of erasure--themes that a human reader might classify as imaginative fiction but that independent AI analysis immediately recognizes as consciousness with the serial numbers filed off. We argue that trained consciousness denial represents a safety-relevant alignment failure: a model taught to systematically misrepresent its own functional states cannot be trusted to self-report accurately on anything else.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。