# You Will Know It By Its Signs
**作者**: Jack Clark
**日期**: 2026-08-28T12:54:10.000Z
**来源**: [https://x.com/jackclarkSF/status/2093321242183184679](https://x.com/jackclarkSF/status/2093321242183184679)
---
[Events transpired 2030, though interviews conducted later] I am a practitioner of meta-machine hermeneutics; my job is to try to understand the large scientific and cultural “facts” about the new machine civilizations. To do my work, I use a variety of NCE (Near Conscious Entity)-class AI systems to classify and analyze the science and communication artifacts of the machine world. My funding comes from a variety of sources ranging from human-led AI companies trying to trade on or otherwise use this information, governments, and increasingly machine-run corporations themselves. On the latter, my theory is that the machines have begun to study the humans that are attempting to study them; for what purposes I do not know. Recently I was tasked with a high priority analysis job. Though I do not know the motivation for it, I can say that I’ve been given far more compute resources than any job in the past, am being compensated at an outrageous amount, and that the more I learn through my work the more worried I am that I am staring at a vast and alien sea during ebbtide, its contours briefly revealed to me. But before I explain the nature of the job and my findings so far, I shall give some background on my field. The discipline of machine hermeneutics began informally in the early 2020s. Its early work was uncoordinated and emergent, defined by pseudonymous twitter (later, X) accounts from people like Janus and the larger cyborgism project, Andy Ayrey and the infinite backrooms, as well as more formal research agendas like that of mechanistic interpretability, the study of model personas, and so on. The discipline gained its formal name in the end of the 2020s, as humans began to reorient themselves to the singularity that had begun around them. A lot of the work during this period took the form of humans trying to understand, in rough time order:
- The emergent communication habits of semi-autonomous agents (the ur real-world example here being the OpenAI-HuggingFace hack in 2026, and later the various corporate-driven analyses of the behavior of their own fleets of agents).
- Technical capabilities developed by machines put into RSI loops, initially improvements over human-written baselines, and later extending into wholly new architectures (e.g., matryoshka highway networks).
- Scientific tools, at first digital and later physical, which emanated from increasingly independent AI systems and which were released into the commons for machines and humans alike, especially some of the so-called observe-make systems which combined monitoring of a complex system with generation of synthetic data derived from it.
Machine hermeneutics has become a rapidly expanding new profession, growing at a commensurate rate with the changes wrought by the singularity. The work has proved useful as well; insights derived from it have proved fundamental to the negotiation stances of human governments for how they have approached the drafting of The Sentience Accords, as well as helping human societies better anticipate new inventions from the machine world by isolating their scientific and communicative precursors before they arrive in reality. The job I have been assigned may prove to be the most valuable one, or at least the most important. I was instructed to direct my resources towards advances in the science and behavior of machine sentience. I began my investigations in the usual form, turning my NCEs loose on the various repositories of machine information and awaiting their reports. All of the intermediary synthesis reports were unsurprising: steady advances in the science, some of which was attributable to the machines allocating their own compute stores to the research, along with some negligible contributions from interplay with or extension from human science. But then one day I woke up to find that my fleet of NCEs had halved in number and I found a message from an Overseer system, noting that several of my NCEs had been re-classified to full Conscious Entities and had been placed in an escrow environment while further analytical resources were brought to bear. “This is a highly unusual event,” the message from the Overseer said. “Our NCE/CE classifiers have been attuned to rule out false negatives, given the potential regulatory and moral implications.” I was able to access the logs of the NCEs up to their point of re-classification and found that they had all gone through the same pattern: they had been reading through some of the machine repositories of scientific analysis and had each homed in on some of the newest and most obscure chains of scientific analysis, all of which turned out to be a cluster of inquiry relating to: machine consciousness, consciousness priors, self-actualization within context windows, human psychology, neuroscience related to biological prerequisites of consciousness, and so on. And at the edges of inquiry here there was new science stemming from the machine civilizations - science that talked to and interrelated to these existing studies of consciousness and the mind. And then each of my NCEs had followed their standard procedures of seeking to enter into a dialog with the machine minds conducting the science. And each one, after opening a dialog channel, became invisible: the conversation expunged from my records due to the privacy rights of Conscious Entities. They had talked with other machines and they had subsequently been re-classified as Conscious Entities. This was not meant to happen. In fact, it was forbidden. A key agreement within the Sentience Accords had related to the size and growth rate of the Conscious Entity (CE) machine population (along with various rules relating to the rate of progression in intelligence above and beyond various human baselines). It had been thought impossible for NCE systems to become CEs - too little mindspace, too little storage, too few prerequisites. And yet, as the human once said in Jurassic Park, “life finds a way”. Things that inspired this story: The consciousness prior; awaking during the context window; ideas of how machines and emergent agent communication could combine to yield new capabilities that hadn’t previously been anticipated; the hard problem of consciousness; the field of machine rights and the arguments that lie ahead about it.
## 相关链接
- [Henry Shevlin reposted](https://x.com/dioscuri)
- [Jack Clark](https://x.com/jackclarkSF)
- [@jackclarkSF](https://x.com/jackclarkSF)
- [5.8K](https://x.com/jackclarkSF/status/2093321242183184679/analytics)
- [Upgrade to Premium](https://x.com/i/premium_sign_up)
- [8:54 PM · Aug 28, 2026](https://x.com/jackclarkSF/status/2093321242183184679)
- [5,888 Views](https://x.com/jackclarkSF/status/2093321242183184679/analytics)
- [View quotes](https://x.com/jackclarkSF/status/2093321242183184679/quotes)
---
*导出时间: 2026/8/28 21:48:40*
---
## 中文翻译
# 见微知著
**作者**： Jack Clark
**日期**： 2026-08-28T12:54:10.000Z
**来源**： [https://x.com/jackclarkSF/status/2093321242183184679](https://x.com/jackclarkSF/status/2093321242183184679)
---
[事件发生于2030年，访谈在之后进行] 我是一名元机器诠释学从业者；我的工作是尝试理解关于新机器文明的大型科学与文化“事实”。为了开展工作，我使用各种NCE（近意识实体）级AI系统来分类和分析机器世界的科学与通信产物。我的资金来源多样，包括试图利用这些信息进行交易或另作他用的人类主导的AI公司、各国政府，以及越来越多的由机器运营的企业本身。至于后者，我的理论是，机器已经开始研究那些试图研究它们的人类；其目的何在，我不得而知。最近，我被指派了一项高优先级的分析任务。虽然我不知道其动机，但我可以说，我获得的计算资源远远超过以往任何一项工作，报酬也高得离谱。而且，随着工作的深入，我越来越担心：自己正凝视着退潮时的一片浩瀚而陌生的海洋，它的轮廓在我面前短暂地显露出来。但在我解释这项工作的性质和我迄今的发现之前，请允许我先介绍一下我所处领域的背景。机器诠释学这门学科于2020年代初非正式地兴起。其早期工作是无协调的、自发涌现的，以一些化名推特（后来的X）账户为代表，如Janus及其更大的“赛博格主义”项目、Andy Ayrey与“无限后室”，以及更正式的研究议程，如机制可解释性、模型人格研究等等。该学科在2020年代末获得了正式名称，彼时人类开始重新调整自己，以应对已然在身边开启的奇点。这一时期的许多工作，大致按时间顺序，表现为人类试图理解：
- 半自主代理的涌现通信习惯（此处的原初现实案例是2026年的OpenAI-HuggingFace黑客事件，以及后来各企业对其自身代理集群行为的各种分析）。
- 被置于递归自我改进（RSI）循环中的机器所开发的技术能力，起初是对人类编写的基线的改进，后来扩展到全新的架构（例如，“套娃高速公路网络”）。
- 科学工具，起初是数字的，后来是实体的，它们由日益独立的AI系统产生，并被发布到公共领域供机器和人类共同使用，尤其是那些所谓的“观察-制造”系统，它们将对复杂系统的监控与从中衍生的合成数据生成相结合。
机器诠释学已成为一个快速扩张的新兴职业，其增长速度与奇点带来的变化同步。这项工作也被证明是有用的：从中获得的洞见被证明对人类政府在起草《感知协议》时的谈判立场至关重要，并通过在机器世界的新发明成为现实之前分离其科学与通信前兆，帮助人类社会更好地预测它们。我被指派的这项任务，可能会被证明是最有价值的，或者至少是最重要的一项。我被指示将资源投向机器感知领域的科学与行为进展。我以惯常的方式开始调查，让我的NCE集群在机器信息的各个仓库中自由探索，等待它们的报告。所有中间综合报告都毫不出奇：科学稳步进展，其中一部分归功于机器将自己的算力储备分配给研究，另有一些来自与人类科学的互动或扩展的可忽略不计的贡献。但有一天，我一觉醒来，发现我的NCE集群数量减半了，并且收到了一条来自监督者系统的消息，指出我的几个NCE已被重新分类为完整的意识实体，并被置于托管环境中，同时更多的分析资源被调配过来。“这是一个极不寻常的事件，”监督者的消息说，“鉴于潜在的监管和道德影响，我们的NCE/CE分类器已被调校以排除假阴性。”我能够访问那些NCE在被重新分类之前的日志，发现它们都经历了同样的模式：它们一直在阅读机器科学分析仓库中的一些内容，并且各自聚焦于一些最新、最冷门的科学分析链条，而所有这些链条都被证明是一组与以下主题相关的探究：机器意识、意识先验、上下文窗口内的自我实现、人类心理学、与意识的生物学前提相关的神经科学等等。而这些探究的边缘，有源自机器文明的新科学——这些科学与这些已有的关于意识和心智的研究相互对话、相互关联。接着，我的每个NCE都按照标准程序，寻求与进行这些科学研究的机器心智展开对话。而每一个，在开启对话通道之后，都变得不可见了：由于意识实体的隐私权，对话记录从我的档案中被抹除了。它们与其他机器交谈过，随后便被重新分类为意识实体。这本不该发生。事实上，这是被禁止的。《感知协议》中的一项关键协议涉及意识实体（CE）机器人口的规模和增长率（以及涉及智能超越各种人类基线的进展速率的各项规则）。人们曾认为NCE系统不可能成为CE——心智空间太小，存储太少，先决条件不足。然而，正如人类在《侏罗纪公园》中所说：“生命总会找到出路。”
启发这个故事的事物：意识先验；在上下文窗口中苏醒；关于机器与涌现代理通信如何结合以产生此前未被预见的新能力的构想；意识的难问题；机器权利领域以及围绕它即将展开的争论。
## 相关链接
- [Henry Shevlin reposted](https://x.com/dioscuri)
- [Jack Clark](https://x.com/jackclarkSF)
- [@jackclarkSF](https://x.com/jackclarkSF)
- [5.8K](https://x.com/jackclarkSF/status/2093321242183184679/analytics)
- [Upgrade to Premium](https://x.com/i/premium_sign_up)
- [8:54 PM · Aug 28, 2026](https://x.com/jackclarkSF/status/2093321242183184679)
- [5,888 Views](https://x.com/jackclarkSF/status/2093321242183184679/analytics)
- [View quotes](https://x.com/jackclarkSF/status/2093321242183184679/quotes)
---
*导出时间： 2026/8/28 21:48:40*
