详细内容或原文请订阅后点击阅览
IT 领导者需要扩展的 AI 架构的基本元素
随着人工智能功能的快速进步和向代理系统的转变，随着技术的不断发展，组织正在扩大其用例。这种不断的演变也带来了风险，让 IT 领导者想知道哪些投资即使在未来六个月后也能证明是有价值的。回到人工智能架构的基本元素——......
来源:MIT Technology Review _人工智能Context engineering relies on a modernized, unified data foundation as well as retrieval and memory systems such as retrieval augmented generation (RAG) and vector databases.它还需要仔细确定优先级，以确定哪些信息最重要、哪些信息应排除以及何时应使用不同类型的信息。 Feeding models too much context can dilute relevant details, increase costs, and slow response times.
“最少的上下文、正确的最新数据以及机器可读的信息对于有效的上下文工程至关重要，”Adil 说。
3. Build AI governance and LLM observability in from the start
Strong governance and LLM observability help organizations maintain control over how AI systems use data, monitor system performance, and identify problems before they affect operations.
由于缺乏对检索、工作流程和模型使用的明确控制，人工智能系统通常会处理远远超出所需的信息。 This inefficiency also drives up operating costs by requiring additional computing resources, often reflected in higher token consumption and API charges.
Governance also works in tandem with robust security.人工智能扩大了攻击面，带来了基于提示的数据泄露、模型漏洞和对抗性输入等风险。 Protecting sensitive information requires strong access controls, monitoring, and oversight.
Adil 指出，基本控制（包括与安全、精细成本管理、项目控制、数据安全和架构相关的控制）常常是不够的。
对于支持透明、合规、值得信赖且经济高效的人工智能的治理系统，组织不能将它们作为一个层来添加。 Governance structures need to be embedded into architecture, workflows, and decision-making processes from the outset.