AIP overview（AIP 概述）¶
Palantir's Artificial Intelligence Platform (AIP) connects AI with your data and operations. AIP is designed to drive automation across operational processes, providing a comprehensive suite of tools that can be used by everyone in an organization, from developers to frontline users.
AIP's builder tools like AIP Logic, AIP Chatbot Studio (formerly known as AIP Agent Studio), and AIP Evals enable the development of production-ready AI-powered workflows, agents, and functions on top of the Ontology and developer toolchain. Additionally, AIP transforms the application environment by allowing sandboxed, autoscaling applications to integrate generative AI seamlessly within existing security, audit, and resource management frameworks.
Together with Foundry (Palantir's data operations platform) and Apollo (Palantir's mission control for autonomous software deployment), AIP forms an operating system that can deliver a full range of AI-driven products, from LLM-powered web applications to mobile applications using vision-language models to edge applications that embed localized AI.
:::callout{theme="success"} Get started with the "Speedrun: Your first AIP workflow" ↗ course on learn.palantir.com ↗. :::
The remainder of this page provides a brief overview of key AIP benefits. For more details about AIP's capabilities, we recommend reviewing the AIP documentation, including the AIP application reference.
Seamless integration¶
Palantir AIP can integrate seamlessly with your organization's existing data on a Foundry enrollment. This enables you to build and interact with LLM-powered agents and workflows that leverage data from a wide array of data sources and formats.
Security and governance¶
AIP incorporates all of Palantir's advanced security measures for the protection of sensitive data in compliance with industry regulations. AIP provides robust access controls, encryption, and auditing capabilities to maintain data integrity and transparency. Moreover, built-in governance tools help organizations maintain accountability and historical lineage in AI operations.
To learn more about how LLMs securely process user prompts in the platform, review FAQs: Security and Privacy of Palantir's AIP leveraging third-party-hosted LLMs ↗ by selecting Palantir AIP FAQs.
Model management¶
AIP provides a comprehensive suite of tools for building, training, and deploying large language models. Supporting a range of different large language models allows data scientists and engineers to work with their preferred tools and pick the models that are best suited for each use case. Additionally, AIP offers version control and collaboration features, enabling teams to manage models efficiently throughout their lifecycle.
Scalability and performance¶
Designed to handle large-scale data operations, AIP ensures that AI models can be deployed and scaled according to organizational needs. The platform's architecture supports distributed computing, allowing for high-performance processing and real-time analytics, which are crucial for mission-critical applications. AIP also provides granular control of resource use and limitation setting.
To learn more about the tools that enable developers to monitor and optimize the performance of the agents and applications that are built with the platform, review the AIP observability documentation.
Explainability and transparency¶
Trust is critical when building AI workflows for production deployment; for LLMs, trust comes from explainability and transparency, as well as from rigorous evaluations. AIP provides tools for generating detailed audit trails, explanations, and evaluations of model decisions, helping users understand and trust the outcomes. This trust is vital for organizations to safely deploy AI in real-world situations and make informed and ethical decisions based on AI insights.
Note: AIP feature availability is subject to change and may differ between customers.
中文翻译¶
AIP 概述¶
Palantir 的人工智能平台（AIP）将 AI 与您的数据和运营相连接。AIP 旨在推动运营流程的自动化，提供一套全面的工具，可供组织中的所有人使用，从开发者到一线用户。
AIP 的构建工具，如 AIP Logic、AIP Chatbot Studio（原名 AIP Agent Studio）和 AIP Evals，支持在 Ontology（本体论）和开发者工具链之上开发可用于生产环境的 AI 驱动工作流、智能体（Agent）和函数。此外，AIP 通过允许沙盒化、自动扩展的应用程序将生成式 AI 无缝集成到现有的安全、审计和资源管理框架中，从而改变了应用环境。
AIP 与 Foundry（Palantir 的数据运营平台）和 Apollo（Palantir 的自主软件部署任务控制平台）共同构成一个操作系统，能够提供全方位的 AI 驱动产品，从基于大语言模型（LLM）的 Web 应用程序，到使用视觉语言模型的移动应用，再到嵌入本地化 AI 的边缘应用。
:::callout{theme="success"} 从 learn.palantir.com ↗ 上的 “速通：您的第一个 AIP 工作流” ↗ 课程开始入门。 :::
本页其余部分简要概述了 AIP 的主要优势。有关 AIP 功能的更多详细信息，我们建议查阅 AIP 文档，包括 AIP 应用参考。
无缝集成¶
Palantir AIP 可以与您组织在 Foundry 注册（Enrollment）中的现有数据无缝集成。这使得您能够构建和交互由 LLM 驱动的智能体和工作流，这些智能体和工作流可利用来自多种数据源和格式的数据。
安全与治理¶
AIP 融合了 Palantir 所有先进的安全措施，以保护敏感数据并符合行业法规。AIP 提供强大的访问控制、加密和审计能力，以维护数据的完整性和透明度。此外，内置的治理工具帮助组织在 AI 运营中保持问责制和可追溯的历史沿革。
要了解 LLM 如何在平台中安全处理用户提示，请通过选择 Palantir AIP 常见问题解答，查阅 常见问题解答：Palantir AIP 利用第三方托管的 LLM 的安全与隐私 ↗。
模型管理¶
AIP 提供了一套全面的工具，用于构建、训练和部署大语言模型。支持多种不同的大语言模型，使数据科学家和工程师能够使用他们偏好的工具，并为每个用例选择最合适的模型。此外，AIP 还提供版本控制和协作功能，使团队能够在模型整个生命周期内高效管理模型。
可扩展性与性能¶
AIP 专为处理大规模数据操作而设计，确保 AI 模型能够根据组织需求进行部署和扩展。该平台的架构支持分布式计算，可实现高性能处理和实时分析，这对于关键任务应用至关重要。AIP 还提供对资源使用和限制设置的精细控制。
要了解有关使开发者能够监控和优化使用该平台构建的智能体和应用性能的工具的更多信息，请查阅 AIP 可观测性文档。
可解释性与透明度¶
在为生产部署构建 AI 工作流时，信任至关重要；对于 LLM 而言，信任来自可解释性和透明度，以及严格的评估。AIP 提供用于生成详细审计追踪、解释和模型决策评估的工具，帮助用户理解并信任结果。这种信任对于组织在现实世界中安全部署 AI，并基于 AI 洞察做出明智且合乎道德的决策至关重要。
注意：AIP 功能的可用性可能会发生变化，并且可能因客户而异。