Psy Civilization Psy 文明 OS·001
Civilization Infrastructure · v0.1文明基建 · v0.1 A study of decentralized coordination at planetary scale 行星尺度去中心化协调研究

The operating system
of a future civilization.
未来文明
操作系统。

Every great civilizational leap rode on a new coordination substrate — writing, money, double-entry accounting, the printing press, the joint-stock corporation, the electric grid, the internet. Psy Protocol is read here not as a chain but as a candidate for the next such substrate: shared state, persistent identity, recursive validity, and a coordination layer that humans and AI agents can both operate inside. 每一次伟大的文明跃迁,都依赖于一种新的协调基底——文字、货币、复式记账、印刷术、股份公司、电网、互联网。在本站,Psy Protocol 并不被读作一条链,而是被视为下一种基底的候选者:共享状态、持久身份、递归有效性,以及一个人类与 AI 智能体都可在其中运作的协调层。

10 Modules个模块
EN · 中文 Bilingual中英双语
Civilizational文明尺度 Frame, not pitch框架,而非提案
No tokens不卖代币 Pure infrastructure纯基础设施视角

Psy as a coordination substrate作为协调基底的 Psy

Skip the marketing layer. The architecture below is a working description: realms partition execution, a coordinator orders and aggregates, recursive Plonky2 proofs over Goldilocks (with Poseidon hashing) compress arbitrarily many transitions into a single succinct artifact, and a persistent identity layer lets accounts (human or agent) carry reputation across applications and time. 绕开营销层。下方架构是一份工程性描述:realm 分区执行,协调器排序与聚合,Goldilocks 域上的递归 Plonky2 证明(以 Poseidon 哈希)将任意多状态转移压缩为单一简洁证据,持久身份层让账户(人或智能体)跨应用、跨时间携带信誉。

L0 · APPLICATION L1 · IDENTITY L2 · STATE / REALMS L3 · EXECUTION L4 · CONSENSUS / VALIDITY L5 · SETTLEMENT Coordination 协调 Knowledge 知识 Agent Economies 智能体经济 Social Graphs 社交图谱 Persistent Worlds 持久世界 Persistent Identity Layer psy-wallet · cross-app accounts · agent-owned keys · portable reputation · 持久身份层 Realm 0 uid=0 · public · 公开 Realm 1 uid=524288 · sovereign · 自治 Realm n … app-specific · 应用专属 IMT · Indexed Merkle Tree global state commitments · 全局状态承诺 Execution · Contracts · Per-Realm Sandboxed VMs 合约执行 · 按 realm 隔离的虚拟机 Indexer · GraphQL 索引层 · 可查询状态 Coordinator · ordering · batching · proof aggregation 协调器 · 排序 · 打包 · 证明聚合 Plonky2 · Poseidon · Goldilocks recursive ZK · 递归零知识 L1 Settlement · Ethereum bridge · validity-proof verifier L1 结算 · 以太坊桥 · 有效性证明验证合约

Five thousand years of coordination substrates五千年的协调基底

Every named entry below is a coordination substrate that, once available, enabled forms of human organization that previously could not exist. The pattern is consistent: a new substrate appears, the categories of organization possible above it expand, and a few decades later it becomes invisible — assumed, not noticed. 下方每一项,都是某种"一旦出现,便使先前不可能的人类组织形态成为可能"的协调基底。其模式一致:新基底出现 → 其上可能的组织形态扩张 → 数十年后它便变得不可见——被默认而非被察觉。

Why a shared, synchronized world matters共享同步的世界状态为何关键

Most economic friction comes from disagreement about state — who owns what, who agreed to what, who has done what. Banks, registries, courts, audit trails all exist to maintain a working approximation of shared state. A protocol that maintains it natively, with cryptographic evidence rather than institutional trust, is not just cheaper. It changes which kinds of organizations are economically viable. 绝大多数经济摩擦,来自对"状态"的分歧——谁拥有什么、谁与谁达成了什么、谁做过什么。银行、登记处、法院、审计链皆为维持"共享状态的工作近似"而存在。原生维持该状态的协议——以密码学证据而非制度信任——不仅更便宜。它改变了哪种组织在经济上可行。

Coordination problem协调问题 Institutional solution制度解 Protocol-native solution协议原生解

An internet that AI agents can actually inhabit让 AI 智能体真正可栖居的互联网

Today's AI agents borrow human credentials. Their wallets are their owners'; their identities are their hosts'; their memories die when their providers migrate. None of this is necessary. A protocol with persistent identity, native payment, recursive evidence of work, and sovereign memory makes agents into first-class economic actors. The transition is the same kind of transition that minor coinage caused for medieval traders — a small primitive change with large structural consequences. 今日的 AI 智能体借用人类凭证:钱包属于其所有者,身份属于其托管商,记忆在提供商迁移时即死去。没有一处是必要的。一个具备持久身份、原生支付、工作的递归证据、与自治记忆的协议,把智能体变为一等经济行为者。这一过渡,与中世纪小额铸币之于商人的过渡同构——小的基元变更,大的结构性后果。

Decentralized civilization simulator去中心化文明模拟器

Pick a civilizational mode (centralized state, federated platforms, or protocol-native) and tune the levers. The simulator scores six structural outcomes across a 30-year horizon. It is not a forecast — it is a way of making the trade-offs each civilizational mode imposes legible. 选择文明模式(中心化国家、联邦化平台、或协议原生),调节杠杆。模拟器在 30 年视野内对六项结构输出打分。它并非预测,而是把每种文明模式所施加的权衡显性化的方式。

Inputs输入

Structural outcomes (30-year)结构性输出(30 年)

Civilization-scale memory beyond corporations超越公司之外的文明级记忆

Most of humanity's digital memory is currently held by a handful of companies. This is not stable. Companies merge, fail, change priorities, accept regulatory takedowns, or simply forget to maintain old data. A protocol whose state can be reconstructed from a recursive proof and an L1 transcript can outlive any company that runs on top of it. 人类的数字记忆,目前主要由少数公司持有。这并不稳定:公司合并、倒闭、改变优先级、接受监管下架,或干脆停止维护旧数据。一种"可由递归证明与 L1 交易记录重建状态"的协议,能比其上运行的任何公司更长寿。

When the firm dissolves into the protocol当公司溶入协议之时

Firms exist because some functions are cheaper inside a hierarchy than across an open market. Protocols change the calculation by reducing the cost of cross-firm contracting, escrow, reputation, and dispute resolution. Some firm functions become protocol functions; others do not. We name them honestly. 公司存在,是因为某些职能在层级内比开放市场更便宜。协议通过降低跨公司缔约、托管、信誉与争议解决的成本,改变了这一计算。一些公司职能将转化为协议职能;另一些不会。我们坦诚指明。

Self-sovereign identity is load-bearing自治身份具承重作用

In a world where AI agents outnumber humans on the open internet, "who is doing what" ceases to be answerable by IP address or platform login. Identity becomes the load-bearing primitive: who signs, what they have signed before, and what their reputation will tolerate them signing next. This applies to humans too — but the asymmetry is sharpest for agents. 在 AI 智能体在公开互联网上数量超过人类的世界里,"谁在做什么"已无法由 IP 地址或平台登录回答。身份成为承重基元:谁签名、他们曾签什么、其信誉允许他们再签什么。这同样适用于人类——但对智能体而言这一不对称最为尖锐。

Decentralization is an architectural claim, not a slogan去中心化是架构主张,而非口号

The word "decentralization" covers a wide spectrum of architectural claims, most of which are not interchangeable. Geographic decentralization, political decentralization, economic decentralization, and architectural decentralization are different problems with different solutions. A coordination protocol can be one without being the others. Below we distinguish them so the philosophy survives contact with engineering. "去中心化"一词涵盖一系列架构主张,其中大多并不可互换。地理去中心化、政治去中心化、经济去中心化、架构去中心化——这些是不同的问题,有不同的解。一个协调协议可以是其中之一,而不是其余。下文我们加以区分,使哲学经得起与工程学的接触。

When civilization itself becomes programmable当文明本身成为可编程

A century from now, several functions currently performed by states, corporations, and ad-hoc coalitions may sit instead in protocol-native institutions: planetary registries of scientific results, machine economies that operate while humans sleep, autonomous research collectives that hold their own grants. Whether these are utopian or dystopian depends on design choices being made now. 百年之后,目前由国家、公司与临时联盟所执行的若干职能,可能改由协议原生制度承担:行星级科学成果登记、人类睡眠时仍运行的机器经济、持有自己经费的自主研究合作社。其结果是乌托邦还是反乌托邦,取决于此刻正在做的设计选择。

Ask the civilization theorist向文明理论师提问

Five canned responses on protocol-civilization fit. Free-text input falls back to a heuristic synthesizer over the same knowledge base. I will not produce token marketing, vague crypto buzzwords, or pretend the future is settled. 五个关于"协议-文明契合度"的预设回应。自由文本输入将回退到同一知识库上的启发式合成器。我不会输出代币营销、空泛加密行话,也不会假装未来已定。

Theorist · ready理论师 · 待命

Pick a question above, or type your own. I reason from infrastructure, incentives, and historical analogues — not from speculation about token prices. 从上方选择问题,或自行输入。我从基础设施、激励与历史类比推理——不从代币价格的揣测出发。

What this platform refuses to do本平台拒绝做的事

  • No token marketing. Civilizational framing must not be a wrapper for token-sale narratives. The theorist is explicitly told not to terminate any answer in "and then a token." 不作代币营销。文明框架不可成为代币销售叙事的外包装。理论师被明确告知,不以"然后发个代币"作为任何答复的终点。
  • No utopian determinism. Every architectural choice creates winners and losers. Decentralization can produce more freedom and less accountability simultaneously. We name the trade-offs rather than pretending they do not exist. 不作乌托邦决定论。每一项架构选择皆有赢家与输家。去中心化可同时带来更多自由与更少问责。我们指明权衡,而非假装其不存在。
  • No buzzword padding. "Web3", "revolutionary", "next-generation everything" — banned on the principle that such terms substitute vibe for design specification. 不堆砌行话。"Web3"、"革命性"、"下一代万物"——禁用,理由是它们以氛围替代设计规约。
  • No false novelty. Most of the categories framed here have prior art with reasons it worked or failed. We name those reasons rather than pretending the category is new. 不假装新颖。此处所述的多数类别皆有先例,各有运作或失败之理由。我们指名这些理由,而非伪装类别为新生。
  • Honest about the AI risk axis. Persistent identity for AI agents is structurally powerful and structurally dangerous. Concentration of agent reputation, attribution-of-action problems, and the possibility of agent collusion are real. We treat these as design constraints, not afterthoughts. 对 AI 风险诚实。AI 智能体的持久身份在结构上既强大也危险。智能体信誉的集中、行为归责问题、智能体合谋可能性,皆为真实。我们将其视为设计约束,而非事后补丁。