Why create an Ontology?（为什么要创建本体论(Ontology)？）

The Palantir platform powers real-time, human-agent decision-making in the most critical commercial and government contexts around the world. The Ontology is the central system that enables customers to safely, securely, and effectively leverage AI in their enterprises and drive operational transformation.

The Ontology represents the decisions in an enterprise, not simply the data. With the Ontology, organizations can make the best possible decisions, often in real time, based on constantly changing internal and external conditions. Traditional data architectures do not capture the reasoning that goes into decision-making or the actions that follow, and therefore limit learning and the incorporation of AI. Conventional analytics architectures do not contextualize computation in lived reality, and remain disconnected from operations. In contrast, the decision-centric Ontology connects humans and agents directly to operations to face and overcome an organization's toughest challenges.

Understanding the value of the Ontology

Palantir models each operational decision as comprising four components:

• Data: The information leveraged to make the decision.
• Logic: The heuristics and computational processes that evaluate a decision.
• Action: The orchestration and execution of the chosen decision.
• Security: The assurance that the decision complies with operational policies.

The Ontology integrates these four elements into a scalable, dynamic, collaborative resource which enables decision-making, informed by the ever-changing conditions and needs of your organization.

Data

The Ontology includes not only the many sources of enterprise data — structured data, streaming and edge sources, unstructured repositories, imagery data, and more — but also the data generated by end users and agents as decisions are being made. This "decision data" contains the context surrounding a given decision, the different options evaluated, and the downstream implications of the committed choice. Integrating the full range of enterprise data alongside decision data requires a different architecture than a classical database management solution optimized for reporting and analytics.

The Ontology integrates this data into a full-scale, full-fidelity semantic representation of the enterprise. The wide range of operational data sources (such as ERPs, MES, and WMS) can be synchronized and contextualized alongside data streams from IoT and edge systems, the relevant sections of unstructured data repositories, geospatial data stores, and more. The Ontology unites these data sources in the form of objects, properties, and links which evolve in real-time, and are designed to be embedded directly into decision-making workflows.

The Ontology safely captures the decision data produced by operational users as they carry out daily work, whether in supply chains, hospital systems, customer service centers, or elsewhere. This includes decisions made at the edge, captured through the lightweight Embedded Ontology ↗. The end-to-end "decision lineage" of when a given decision was made, atop which version of enterprise data, and through which application, is automatically captured and securely accessible to both human developers and agents. Together, these data resources can power AI-driven learning at scale and continuously refine short-term and long-term agentic memory.

Logic

The data stored in the Ontology is complemented by the reasoning, or logic, that determines when and how to make a given decision. Examples of decision logic include a simple piece of business logic within a core business system, a forecast model maintained using a cloud data science workbench, or an optimization model that uses several data sources to produce an operational plan.

With the advent of agentic orchestration, it is critical that AI-driven reasoning can leverage these logical assets in the same way that humans have historically. Deterministic functions, algorithms, and conventional statistical processes can serve as operational tools which complement the non-deterministic reasoning of LLMs and multi-modal models.

The Ontology enables the full set of logic assets — the calculations and processes that dictate how decisions are made — to be connected and contextualized for both human and agents. This includes business logic related to customer interactions, often found in CRMs and ERPs; the modeling logic that drives conventional machine learning, which is spread across data science environments; and the planning, optimization, and simulation algorithms that are often associated with domain-specific tools.

The Ontology’s flexible "logic binding" paradigm provides a consistent interface for constructing workflows that incorporate and combine heterogeneous logic assets from different environments (such as on-premises data centers, enterprise cloud environments, SaaS environments, or the Palantir platform itself). This enables the introduction of agent-driven reasoning into decision-making contexts with diverse sets of logic, which were previously the exclusive domain of human users.

Action

With both information (the data) and reasoning (the logic) incorporated into a shared representation, the next piece is the execution and orchestration of the decision itself (the action). Closing the action loop as decisions are made in real-time is what distinguishes an operational system from an analytical system.

The Ontology natively models actions within a cohesive, decision-centric model of the enterprise. If the data elements in the Ontology are “the nouns” of the enterprise (the semantic, real-world objects and links), then the actions can be considered “the verbs” (the kinetic, real-world execution). With every Ontology-driven workflow, the nouns and the verbs are brought together into complete sentences through human- and/or AI-driven reasoning, which incorporates various pieces of logic.

Uniting data within a semantic model and combining it with the logic required to evaluate decisions is valuable, but ultimately limited unless the executed decisions can be synchronized with operational systems in a way that compounds, with each decision informing the next in a shared lineage. The Ontology enables human and agent actions to be safely staged as scenarios, governed with the same granular access controls as data and logic primitives, and securely written back to every enterprise substrate (transactional systems, edge devices, custom applications, and so on).

Security

In an operational setting, human-agent interaction requires rigorous security and governance capabilities that can go beyond conventional role-driven policies on buckets of data. Palantir provides a security architecture that combines:

• Marking-, purpose-, and role-based policies;
• Dynamic lineage that flows across data, logic, action, and application artifacts; and
• A full suite of integrated change and release management tools that apply across both human-driven and agentic workflows.

Granular policies can constrain both agentic and human access to sensitive or context-dependent information across the Ontology. These policies are dynamically computed at runtime for every interaction, combining row- and column-level restrictions that have been applied to underlying datasets, attributes of particular user groups (including those that flow via SSO), security markings that propagate across underlying data pipelines, and more.

Tool usage is dynamically enforced through the same security architecture that governs data access and all forms of memory. This ensures, at minimum, that any tool invocations are dependent on access to the underlying objects, properties, and links in the Ontology. Tools can also contain runtime validations that are dependent on granular submission criteria.

Every agentic or human action depends on precise authorization grants that explicitly dictate the set of allowable operations, safeguarding against unexpected invocations (such as querying data that exists across organizational boundaries, or tools that connect to unspecified external systems) and other forms of privilege escalation.

As detailed telemetry is generated by agents, the security and transmission of the logs is a critical last-mile concern. Palantir enables administrators to control how logging is accessible across specific projects, workflows, and agents. Data markings and other active security primitives govern log access, in the same manner that they govern access to the underlying data, logic, and action primitives.

The Ontology brings together data, logic, action, and security into a decision-centric model of the enterprise, which can be jointly leveraged by both humans and agents. From data integration to application building to end user workflows, the platform's modular architecture enables human users and agents to query, reason, and act across a shared operational foundation.

Example of an operational workflow

This section provides a notional example of how the Ontology can enable human-agent workflows in an organization.

Background

In this scenario, Onyx Incorporated, a fictional manufacturer of medical equipment, produces a range of finished goods, from syringes to surgical masks, each of which requires moving a precise set of materials through an associated manufacturing process. A diverse set of teams manages everything from supplier relations, to warehouse operations, to production of the finished goods, to distribution to end customers; decisions are interdependent, and constantly adapting to changing circumstances.

Imagine that Onyx is faced with an unexpected disruption with one of their major suppliers, which provides the key raw materials needed to produce surgical masks. Given the tight production schedules across Onyx’s manufacturing plants and the escalating demand from customers for surgical masks, this disruption could create serious issues with fulfilling outstanding customer orders. To respond, Onyx’s operational teams have decided to use Palantir's AI FDE to connect a wide array of data sources, logic assets, and systems of action into their enterprise ontology.

Gaining visibility into the problem

Onyx will start by assessing the immediate impact of the supplier shortage, and will then employ AI to assess possible reallocation strategies across production lines, before finally translating their decisions into a set of connected actions that will simultaneously update warehouse processes, production schedules, and fulfillment routes.

Onyx’s ontology provides real-time, end-to-end visibility into the operations happening across each interdependent part of the business. This enables both leadership and on-the-ground teams to quickly understand the supplier disruption. Vital data systems related to supplier management, warehouse operations, production activity within plants, distribution center processing, and customer fulfillment are all synthesized into semantic objects and links, which reflect the language of the business. Using the Palantir platform, an operations leader can rapidly pinpoint the surgical mask production at risk due to the raw material shortage, and through the connections in their ontology, navigate to every outstanding customer order that is now also at risk. The Ontology’s granular security model ensures that more sensitive data elements (such as financial metrics) are automatically hidden by default, as the response widens to include more teams across the enterprise.

While operational users can easily navigate the Ontology through Workshop- and SDK-driven applications, the inclusion of agentic capabilities is a force multiplier for Onyx Incorporated. Agents leveraging both open-source and proprietary LLMs can navigate across supplier information, stock levels, real-time production metrics, shipping manifests, and customer feedback all contained within the organization’s ontology. Importantly, all agentic activity is controlled with the same security policies that govern human usage, ensuring that Onyx engineers always have precise control over what the LLMs can query, recommend, and act upon. Each constructed and deployed agent can be treated like a new team member that is gradually granted a wider purview as Onyx team members gain confidence in its performance.

Building simulations and designing solutions

Situational awareness is only the tip of the ontological iceberg. Onyx needs to rapidly identify solutions to deal with the supplier disruption, and explore the tradeoffs inherent with each possible decision.

Since the diverse set of forecast models, allocation models, production optimizers, and other logic assets have been connected into Onyx’s ontology (alongside the aforementioned data sources), Onyx supply chain analysts can quickly run a battery of simulations that detail the consequences of the different possible material substitutions. The connected, real-time nature of the Ontology is key at this stage, since substituting raw materials will potentially have downstream implications for the other products (like syringes and gloves) being produced from the same materials. As the simulations are run, the simulated outputs are staged as ontology scenarios, which safely package the proposed changes into a sandboxed subset of the Ontology — enabling teams to safely explore and analyze the implications of the decision before committing to it.

Even more valuable for the Onyx team is that fleets of agents can securely leverage the full range of logic assets and the same scenarios framework. The Ontology enables agents to go beyond the data-centric limitations of retrieval-augmented generation, and instead interface with the interconnected data, logic, and action primitives in the Ontology through an extensible tools paradigm. As Onyx’s analytics and data science teams create new machine learning models in their cloud workbenches, tune optimization algorithms within enterprise systems, and fine-tune LLMs using Palantir’s open model building framework, the Ontology can securely surface these logic assets as AI-ready tools.

In this case, Onyx has created a tuned agent, "Disruption Bot", that can use a set of Ontology-driven tools to scan across the full range of enterprise data sources, the after-action reports on prior courses of action taken in similar situations, and the potentially applicable material reallocation models. Thanks to the rich, dense context provided through the Ontology, Disruption Bot is able to surface a novel reallocation plan, which uses a newer model that the supply chain analysts had not yet considered. With the consequences of the plan safely staged in a scenario, the agent’s proposed decision is handed off to a human analyst for final review.

Executing decisions and taking action

With a viable plan to address the material shortage identified, Onyx needs to rapidly and safely push the decision to the operational systems that run the constituent processes. Given that the enterprise has grown through acquisition, and contains a diverse and delicate mix of critical operational systems, the Onyx IT team is vigilant about which processes can write back to these systems, and under which conditions. Fortunately, the Ontology applies the same rigorous control and validation to actions as it does to data and logic; enabling fine-grain control over who can invoke a given action, test-driven frameworks for publishing changes, the ability to stage and review changes in batch, and detailed logging for every event. In this case, the execution of the material reallocation plan automatically orchestrates a set of writeback routines, each tuned for the receiving system: the warehouse management system receives an API-driven update; the three ERP systems each receive updates via native Ontology-driven connectors, which abide by the safeguards in each system; and the production planning system receives a consolidated flat file, which it ingests asynchronously. As actions are executed, the Onyx IT team can monitor system responses and can always audit past activity.

The Ontology provides the guardrails needed for AI to safely take action within permitted boundaries. Alongside data and logic, actions can be automatically surfaced as tools for all types of agents. The scope of an action can be limited to simply reflecting a given change (such as an edit to an object or the creation of a new object) in the Ontology itself; or can write back to single or multiple systems. In this case, Onyx has granted Disruption Bot and a few other production AI agents access to a small set of actions. In the default case, these actions (like changing the status of a work order or pushing a reallocation plan) can only be staged by the AI, before being handed off to a human for final review. However, with the granular logging and operational instrumentation provided by the Ontology (and the wider Palantir platform), Onyx is able to carefully choose whether any trusted, well-tested AI processes can automatically close the action loop without human review. As conditions evolve, the latitude given to AI can be expanded or contracted, with any change instantly reflected across all Ontology-driven workflows.

Learning from decisions

What comes after the immediate crisis is past? With data, logic, action, and security all connected into Onyx’s ontology, the organization can conduct powerful decision-centric learning. The human-agent teaming that produced a specific solution to the material shortage also revealed generalizable workflows, which the organization will want to memorialize and surface in the future. Every data element, logic asset, and action assessed is captured in an end-to-end decision lineage, which serves as rich, contextual fuel for optimizing the performance of AI. The aggregate decisions made by thousands of users and agents throughout Ontology can be securely leveraged as training data when fine-tuning models, and can be distilled into targeted principles that are called upon during agent prompting. The tribal knowledge that has been traditionally trapped in the seams of workflows can be illuminated by AI to improve the operation of the entire enterprise.

Onward with the Ontology

The Ontology allows organizations to implement and scale human-agent operations, as well as precisely control how and when agent-driven recommendations, augmentations, and automations can be used in frontline contexts. This is possible because the Ontology is decision-centric, not simply data-centric, bringing together the constituent elements of decision-making — data, logic, action, and security — in a single software system.

With the Ontology, new data can be rapidly integrated into a full-fidelity semantic representation; new algorithms and business logic can be seamlessly surfaced for both human and AI users; and robust action integration can be achieved through real-time connections with the full range of operational systems. Each organization’s ontology is a real-time representation of the changing conditions, goals, and decisions being made across teams, which ensures that AI usage remains anchored in the reality of the enterprise.

To learn more, explore our documentation on the Ontology’s underlying decision-centric architecture; the extensibility provided through the Ontology SDK; and the Global Branching framework that allows for safe and zero-downtime evolution of the Ontology.

---

中文翻译

为什么要创建本体论(Ontology)？

Palantir平台为全球最关键的商业和政府场景提供实时的人机协同决策支持。本体论(Ontology)作为核心系统，使客户能够安全、可靠且高效地在企业中运用人工智能，推动运营转型。

本体论(Ontology)代表企业中的决策，而不仅仅是数据。借助本体论(Ontology)，组织可以根据不断变化的内部和外部条件，通常实时地做出最佳决策。传统数据架构无法捕捉决策过程中的推理逻辑或后续行动，因此限制了学习和人工智能的整合。传统分析架构无法将计算置于现实场景中，始终与运营脱节。相比之下，以决策为中心的本体论(Ontology)将人类和智能体直接连接到运营环节，以面对和克服组织最严峻的挑战。

理解本体论(Ontology)的价值

Palantir将每个运营决策建模为四个组成部分：

• 数据(Data)： 用于决策的信息。
• 逻辑(Logic)： 评估决策的启发式方法和计算过程。
• 行动(Action)： 选定决策的编排与执行。
• 安全(Security)： 确保决策符合运营策略。

本体论(Ontology)将这四大要素整合为一个可扩展、动态、协作的资源，使决策能够根据组织不断变化的条件和需求而做出。

数据(Data)

本体论(Ontology)不仅包含企业数据的多种来源——结构化数据、流数据和边缘数据源、非结构化存储库、影像数据等——还包括最终用户和智能体在决策过程中生成的数据。这种"决策数据"包含特定决策的上下文、评估的不同选项以及已执行决策的下游影响。将企业数据的完整范围与决策数据整合在一起，需要不同于传统数据库管理解决方案（专为报告和分析优化）的架构。

本体论(Ontology)将这些数据整合为企业全规模、全保真的语义表示。广泛的运营数据源（如ERP、MES和WMS）可以与来自物联网和边缘系统的数据流、非结构化数据存储库的相关部分、地理空间数据存储等同步并上下文化。本体论(Ontology)以对象、属性和链接的形式统一这些数据源，这些元素实时演变，并设计为直接嵌入决策工作流中。

本体论(Ontology)安全地捕获运营用户在日常工作中产生的决策数据，无论是在供应链、医院系统、客户服务中心还是其他场景。这包括通过轻量级嵌入式本体论(Embedded Ontology) ↗捕获的边缘决策。特定决策何时做出、基于哪个版本的企业数据、通过哪个应用程序做出的端到端"决策谱系"会被自动捕获，并且人类开发者和智能体都可以安全访问。这些数据资源共同支持大规模AI驱动的学习，并持续优化短期和长期的智能体记忆。

逻辑(Logic)

本体论(Ontology)中存储的数据由推理或逻辑补充，这些逻辑决定了何时以及如何做出特定决策。决策逻辑的示例包括核心业务系统中的简单业务逻辑、使用云数据科学工作台维护的预测模型，或使用多个数据源生成运营计划的优化模型。

随着智能体编排的出现，AI驱动的推理能够像人类历史上所做的那样利用这些逻辑资产，这一点至关重要。确定性函数、算法和传统统计过程可以作为运营工具，补充大语言模型和多模态模型的非确定性推理。

本体论(Ontology)使完整的逻辑资产集——决定如何做出决策的计算和过程——能够为人类和智能体连接和上下文化。这包括与客户交互相关的业务逻辑（通常存在于CRM和ERP中）；驱动传统机器学习的建模逻辑（分布在数据科学环境中）；以及通常与特定领域工具相关的规划、优化和模拟算法。

本体论(Ontology)灵活的"逻辑绑定"范式提供了一致的接口，用于构建整合来自不同环境（如本地数据中心、企业云环境、SaaS环境或Palantir平台本身）的异构逻辑资产的工作流。这使得智能体驱动的推理能够引入到具有多样化逻辑集的决策场景中，而这些场景以前是人类的专属领域。

行动(Action)

在将信息（数据）和推理（逻辑）整合到共享表示中之后，下一个部分是决策本身的执行和编排（行动）。在实时决策时闭环行动循环，是运营系统与分析系统的区别所在。

本体论(Ontology)在企业统一、以决策为中心的模型中原生建模行动。如果本体论(Ontology)中的数据元素是企业的"名词"（语义化的现实世界对象和链接），那么行动可以被视为"动词"（动态的现实世界执行）。在每个由本体论(Ontology)驱动的工作流中，名词和动词通过人类和/或AI驱动的推理（结合各种逻辑片段）组合成完整的句子。

将数据统一在语义模型中，并结合评估决策所需的逻辑是有价值的，但除非执行的决策能够以复合的方式与运营系统同步——每个决策在共享谱系中为下一个决策提供信息——否则最终是有限的。本体论(Ontology)使人类和智能体的行动能够安全地作为场景暂存，使用与数据和逻辑原语相同的细粒度访问控制进行管理，并安全地写回每个企业基础系统（事务系统、边缘设备、自定义应用程序等）。

安全(Security)

在运营环境中，人机交互需要严格的安全和治理能力，这些能力可以超越传统基于角色的数据桶策略。Palantir提供了一种安全架构，结合了：

• 基于标记、目的和角色的策略；
• 跨数据、逻辑、行动和应用程序工件流动的动态谱系；以及
• 一套完整的集成变更和发布管理工具，适用于人类驱动和智能体驱动的工作流。

细粒度策略可以限制智能体和人类对本体论(Ontology)中敏感或上下文相关信息的访问。这些策略在每次交互时动态计算，结合了应用于底层数据集的行级和列级限制、特定用户组的属性（包括通过SSO流动的属性）、跨底层数据管道传播的安全标记等。

工具使用通过管理数据访问和所有形式记忆的同一安全架构动态执行。这至少确保任何工具调用都依赖于对本体论(Ontology)中底层对象、属性和链接的访问。工具还可以包含依赖于细粒度提交标准的运行时验证。

每个智能体或人类行动都依赖于精确的授权许可，明确指定允许的操作集，防止意外调用（如查询跨组织边界存在的数据，或连接到未指定外部系统的工具）和其他形式的权限提升。

随着智能体生成详细的遥测数据，日志的安全性和传输是关键的最后环节问题。Palantir使管理员能够控制日志在特定项目、工作流和智能体之间的可访问性。数据标记和其他主动安全原语管理日志访问，方式与管理底层数据、逻辑和行动原语的访问相同。

本体论(Ontology)将数据、逻辑、行动和安全整合到以决策为中心的企业模型中，人类和智能体可以共同利用该模型。从数据集成到应用程序构建再到最终用户工作流，平台的模块化架构使人类用户和智能体能够在共享的运营基础上进行查询、推理和行动。

运营工作流示例

本节提供了一个概念性示例，说明本体论(Ontology)如何在组织中实现人机工作流。

背景

在此场景中，虚构的医疗设备制造商Onyx Incorporated生产一系列成品，从注射器到外科口罩，每种产品都需要通过相关的制造流程移动精确的一组材料。多样化的团队管理从供应商关系到仓库运营、成品生产再到最终客户分销的一切事务；决策相互依赖，并不断适应变化的情况。

假设Onyx面临一个主要供应商的意外中断，该供应商提供生产外科口罩所需的关键原材料。考虑到Onyx各制造工厂紧张的生产计划以及客户对外科口罩不断增长的需求，这种中断可能会在履行未完成的客户订单时造成严重问题。为了应对，Onyx的运营团队决定使用Palantir的AI FDE将广泛的数据源、逻辑资产和行动系统连接到他们的企业本体论(Ontology)中。

获取问题的可见性

Onyx将首先评估供应商短缺的直接影响，然后使用AI评估跨生产线可能的重新分配策略，最后将决策转化为一组关联行动，同时更新仓库流程、生产计划和履行路线。

Onyx的本体论(Ontology)提供对业务每个相互依赖部分运营的实时端到端可见性。这使得领导层和一线团队都能快速了解供应商中断情况。与供应商管理、仓库运营、工厂内生产活动、配送中心处理和客户履行相关的重要数据系统都被综合为语义对象和链接，反映业务语言。使用Palantir平台，运营领导者可以快速定位因原材料短缺而面临风险的外科口罩生产，并通过本体论(Ontology)中的连接，导航到每个现在也面临风险的未完成客户订单。本体论(Ontology)的细粒度安全模型确保更敏感的数据元素（如财务指标）在响应范围扩大到包括企业更多团队时默认自动隐藏。

虽然运营用户可以通过Workshop和SDK驱动的应用程序轻松导航本体论(Ontology)，但智能体能力的引入对Onyx Incorporated来说是力量倍增器。利用开源和专有大语言模型的智能体可以导航组织本体论(Ontology)中包含的供应商信息、库存水平、实时生产指标、运输清单和客户反馈。重要的是，所有智能体活动都受管理人类使用的相同安全策略控制，确保Onyx工程师始终精确控制大语言模型可以查询、建议和采取行动的内容。每个构建和部署的智能体都可以被视为一个新团队成员，随着Onyx团队成员对其性能建立信心，逐步获得更广泛的权限。

构建模拟和设计解决方案

态势感知只是本体论(Ontology)价值的冰山一角。Onyx需要快速识别应对供应商中断的解决方案，并探索每个可能决策固有的权衡。

由于多样化的预测模型、分配模型、生产优化器和其他逻辑资产已连接到Onyx的本体论(Ontology)中（与前述数据源一起），Onyx的供应链分析师可以快速运行一系列模拟，详细说明不同可能的材料替代方案的后果。本体论(Ontology)的实时连接特性在此阶段至关重要，因为替代原材料可能会对使用相同材料生产的其他产品（如注射器和手套）产生下游影响。运行模拟时，模拟输出被暂存为本体论(Ontology)场景，将提议的更改安全地打包到本体论(Ontology)的沙箱子集中——使团队能够在提交决策前安全地探索和分析决策的影响。

对Onyx团队更有价值的是，智能体群可以安全地利用完整的逻辑资产集和相同的场景框架。本体论(Ontology)使智能体能够超越检索增强生成的数据中心限制，而是通过可扩展的工具范式与本体的互联数据、逻辑和行动原语交互。随着Onyx的分析和数据科学团队在云工作台中创建新的机器学习模型、在企业系统中调整优化算法、以及使用Palantir的开放模型构建框架微调大语言模型，本体论(Ontology)可以安全地将这些逻辑资产作为AI就绪工具呈现。

在这种情况下，Onyx创建了一个调优的智能体"中断机器人(Disruption Bot)"，它可以使用一组由本体论(Ontology)驱动的工具扫描所有企业数据源、类似情况下先前行动方案的行动后报告以及可能适用的材料重新分配模型。得益于本体论(Ontology)提供的丰富、密集的上下文，中断机器人(Disruption Bot)能够提出一个新颖的重新分配计划，该计划使用了供应链分析师尚未考虑的新模型。由于计划的后果安全地暂存在场景中，智能体提出的决策被移交给人类分析师进行最终审查。

执行决策和采取行动

有了应对材料短缺的可行计划，Onyx需要快速且安全地将决策推送到运行组成流程的运营系统。鉴于企业通过收购发展壮大，包含多样且敏感的关键运营系统组合，Onyx的IT团队对哪些流程可以写回这些系统以及在什么条件下写回保持警惕。幸运的是，本体论(Ontology)对行动应用与数据和逻辑相同的严格控制和验证；实现对谁可以调用特定行动的细粒度控制、用于发布更改的测试驱动框架、批量暂存和审查更改的能力，以及每个事件的详细日志记录。在这种情况下，材料重新分配计划的执行自动编排一组写回例程，每个例程针对接收系统进行调整：仓库管理系统接收API驱动的更新；三个ERP系统各自通过原生本体论(Ontology)驱动的连接器接收更新，遵守每个系统中的安全措施；生产计划系统接收合并的平面文件，异步摄取。随着行动的执行，Onyx的IT团队可以监控系统响应，并始终可以审计过去的活动。

本体论(Ontology)提供了AI在允许边界内安全采取行动所需的护栏。与数据和逻辑一起，行动可以自动作为所有类型智能体的工具呈现。行动的范围可以限制为仅在本体论(Ontology)本身中反映特定更改（如编辑对象或创建新对象）；或者可以写回单个或多个系统。在这种情况下，Onyx已授予中断机器人(Disruption Bot)和其他几个生产AI智能体访问一小部分行动的权限。在默认情况下，这些行动（如更改工单状态或推送重新分配计划）只能由AI暂存，然后移交给人类进行最终审查。然而，借助本体论(Ontology)（以及更广泛的Palantir平台）提供的细粒度日志记录和运营仪表化，Onyx能够仔细选择是否任何受信任、经过充分测试的AI流程可以自动闭环行动循环而无需人工审查。随着条件的变化，给予AI的自由度可以扩大或缩小，任何更改都会立即反映在所有由本体论(Ontology)驱动的工作流中。

从决策中学习

当眼前的危机过去之后呢？随着数据、逻辑、行动和安全都连接到Onyx的本体论(Ontology)中，组织可以进行强大的以决策为中心的学习。产生特定材料短缺解决方案的人机协作也揭示了可推广的工作流，组织希望在未来记录和呈现这些工作流。每个评估过的数据元素、逻辑资产和行动都被捕获在端到端的决策谱系中，作为优化AI性能的丰富、上下文燃料。整个本体论(Ontology)中数千用户和智能体做出的聚合决策可以在微调模型时安全地用作训练数据，并可以提炼为在智能体提示时调用的针对性原则。传统上被困在工作流缝隙中的隐性知识可以通过AI揭示，以改善整个企业的运营。

继续使用本体论(Ontology)

本体论(Ontology)使组织能够实施和扩展人机运营，并精确控制如何以及何时在前线场景中使用智能体驱动的建议、增强和自动化。这是可能的，因为本体论(Ontology)是以决策为中心的，而不仅仅是数据中心化的，它将决策的组成要素——数据、逻辑、行动和安全——整合在一个软件系统中。

借助本体论(Ontology)，新数据可以快速集成到全保真语义表示中；新算法和业务逻辑可以无缝呈现给人类和AI用户；通过与所有运营系统的实时连接，可以实现强大的行动集成。每个组织的本体论(Ontology)都是跨团队不断变化的条件、目标和决策的实时表示，确保AI的使用始终锚定在企业现实中。

要了解更多信息，请探索我们关于本体论(Ontology)底层以决策为中心的架构的文档；通过本体论SDK(Ontology SDK)提供的可扩展性；以及允许本体论(Ontology)安全且零停机演进的全局分支(Global Branching)框架。