Skip to main content

Radar > Topics > AI & ML雷达 > 主题 > AI与机器学习

The Case Against Building Your Own Agent Platform反对自建智能体平台的理由

Pete Johnson

By Pete Johnson June 17, 2026 • 13 minute read

You know the meeting. The board wants an AI agent strategy by end of quarter. Someone on the leadership team has read a McKinsey report. You’ve been voluntold to build the platform. The slide deck says “AI-native.” The acceptance criteria are vague. Somebody mentions LangGraph, and somebody else says, “We’ll just wrap it ourselves.”你懂的。董事会要求在季度末之前制定AI智能体战略。领导团队中有人读过麦肯锡的报告。你被“自愿”指派去构建平台。幻灯片上写着“AI原生”。验收标准含糊不清。有人提到LangGraph，另一个人说：“我们自己封装一下就行。”

You ask what “done” looks like. Nobody in the room can answer.你问“完成”是什么样子。房间里没人能回答。

The cost of building this is almost always estimated before anyone has a clear picture of what “this” actually is. And that’s the problem I want to work through here, because the scope of the work being casually assigned to internal platform teams right now is genuinely larger than the people assigning it understand.构建的成本几乎总是在任何人清楚“这”到底是什么之前就被估算出来了。而这正是我想在这里探讨的问题，因为目前被随意分配给内部平台团队的工作范围，实际上比分配任务的人所理解的要大得多。

Build versus buy, flipped in a year自建与购买，一年内逆转

This particular pendulum has swung before. App servers in the late 1990s. Content management systems in the 2000s. Container orchestration in the 2010s. The pattern rhymes every time: When a category is new, the components look deceptively simple. Early adopters build their own. The market catches up. Within 18 months, building becomes the expensive path. Within 36 months, the teams that built internally are rewriting on top of the category winner that emerged while they weren’t looking.这个钟摆以前也摆动过。20世纪90年代末的应用服务器。21世纪初的内容管理系统。2010年代的容器编排。模式每次都相似：当一个新的类别出现时，其组件看起来简单得具有欺骗性。早期采用者自己构建。市场随后跟上。18个月内，自建变成了昂贵的路径。36个月内，内部构建的团队正在基于他们未曾注意到的市场赢家进行重写。

What’s different about the current moment is the speed. Menlo Ventures’ 2025 State of Generative AI in the Enterprise report shows the build-versus-buy split inverted in a single year. In 2024, 47% of enterprise AI solutions were built internally. By late 2025, that number had collapsed to 24%. The market made the decision in 12 months, which is unusual.当前时刻的不同之处在于速度。Menlo Ventures的《2025年企业生成式AI现状》报告显示，自建与购买的比例在一年内发生了逆转。2024年，47%的企业AI解决方案是内部构建的。到2025年底，这一数字骤降至24%。市场在12个月内做出了决定，这很不寻常。

I’ve lived through enough of these transitions to recognize the shape. What I want to do in this piece is explain why I think the scope of “agent platform” is systematically underestimated right now, and what platform engineers should be asking before they commit to building one.我经历过足够多的这类转变，能够识别出这种模式。在这篇文章中，我想解释为什么我认为“智能体平台”的范围目前被系统性地低估了，以及平台工程师在承诺构建之前应该问些什么。

Most “agent platforms” aren’t大多数“智能体平台”并非如此

A lot of the projects labeled “agent platform” right now are actually workflow systems with an LLM in the loop. That’s a meaningful distinction. As Anthropic pointed out in its “Building Effective Agents” guidance, workflows are systems where LLMs and tools are orchestrated through predefined code paths. Agents are systems where LLMs dynamically direct their own processes and tool usage.目前很多被标记为“智能体平台”的项目，实际上只是带有LLM参与的工作流系统。这是一个有意义的区别。正如Anthropic在其“构建有效智能体”指南中指出的，工作流是通过预定义代码路径编排LLM和工具的系统。而智能体是LLM动态指导自身流程和工具使用的系统。

Most of what enterprises are shipping today sits on the workflow side. That’s fine. Workflows have bounded requirements, tractable testing, and predictable failure modes. If your team is building a workflow system, you might reasonably build it yourselves.目前企业交付的大部分内容都处于工作流一侧。这没问题。工作流有明确的需求、可处理的测试和可预测的故障模式。如果你的团队正在构建一个工作流系统，你完全可以自己构建。

The trap is that teams start building for workflows, then get asked to support agents, and discover the jump isn’t incremental. Agents need memory that survives across sessions. They need evaluation that handles nondeterminism. They need governance that tracks actions, not just outputs. They need orchestration that recovers from failure modes a workflow engine never sees.陷阱在于，团队开始为工作流构建，然后被要求支持智能体，却发现这一跃迁并非渐进式的。智能体需要跨会话持久化的记忆。它们需要处理非确定性的评估。它们需要跟踪行为而不仅仅是输出的治理。它们需要能够从工作流引擎从未见过的故障模式中恢复的编排。

Here’s the thesis I want to put on the table: The decision to build an agent platform almost always underestimates the long tail. Memory, governance, eval, and orchestration aren’t features you add to a workflow engine. They’re separate product bets, each with its own maturity curve, its own vendor landscape, and its own team of specialists who’ve been working on it full-time for 18 months while you’ve been doing something else.以下是我想要提出的论点：构建智能体平台的决定几乎总是低估了长尾问题。记忆、治理、评估和编排不是你可以添加到工作流引擎中的功能。它们是独立的产品赌注，每个都有自己的成熟曲线、自己的供应商格局，以及自己的专家团队，这些专家在你做其他事情的时候已经全职工作了18个月。

Let me walk through them.让我逐一分析。

Memory记忆

The assumption inside most build proposals is that memory is a database problem. You’ll pick a vector store, shove conversation history into it, and retrieve relevant chunks when the agent needs context. Done.大多数构建提案中的假设是，记忆是一个数据库问题。你会选择一个向量存储，把对话历史塞进去，然后在智能体需要上下文时检索相关片段。搞定。

Production memory is three separate systems: episodic, semantic, and procedural, each with different retention and retrieval policies. It’s temporal reasoning that tracks when facts were valid, not just what they were. It’s deduplication, multitenant isolation, and explicit source-of-truth governance.生产级记忆是三个独立的系统：情景记忆、语义记忆和程序记忆，每个都有不同的保留和检索策略。它涉及时间推理，跟踪事实何时有效，而不仅仅是事实是什么。它还包括去重、多租户隔离和明确的真实来源治理。

The signal that this is a separate product category, not a feature: Mem0 raised $24 million across seed and Series A. Letta (formerly MemGPT) raised $10M from Felicis. Zep exists as an independent company with a temporal knowledge graph engine. Mem0’s State of AI Agent Memory 2026 report maps 21 frameworks across three hosting models with measurable benchmark gaps between them. On LongMemEval, Zep scores 15 points higher than Mem0 on temporal queries, which tells you these aren’t interchangeable tools that happen to serve the same market.这表明这是一个独立的产品类别，而非一个功能：Mem0在种子轮和A轮融资中筹集了2400万美元。Letta（前身为MemGPT）从Felicis筹集了1000万美元。Zep作为一家拥有时间知识图谱引擎的独立公司存在。Mem0的《2026年AI智能体记忆现状》报告绘制了跨三种托管模型的21个框架，它们之间存在可衡量的基准差距。在LongMemEval上，Zep在时间查询上比Mem0高出15分，这告诉你这些不是恰好服务于同一市场的可互换工具。

This is the component that platform teams underestimate hardest. Memory sounds like a database problem. It isn’t.这是平台团队最低估的组件。记忆听起来像是一个数据库问题。但它不是。

Governance治理

The assumption is that governance is RBAC plus audit logging. Your agents are services. Services get role-based access controls. You log the tool calls. Compliance is happy.假设是治理等于RBAC加审计日志。你的智能体是服务。服务有基于角色的访问控制。你记录工具调用。合规部门满意了。

Agent governance is something different. It spans action authorization, not just data authorization. It requires decision-chain auditability, where you can reconstruct why the agent did what it did, not just what it did. It needs behavioral drift detection, tiered autonomy, and compliance mapped to agent actions rather than data accesses.智能体治理是另一回事。它涵盖行为授权，而不仅仅是数据授权。它需要决策链可审计性，你可以重建智能体为什么做了某事，而不仅仅是它做了什么。它需要行为漂移检测、分层自主性，以及映射到智能体行为而非数据访问的合规性。

Grant Thornton’s 2026 AI Impact Survey of 950 business executives found that 78% lack strong confidence they could pass an independent AI governance audit within 90 days. Meanwhile, enterprises are moving to increase agent autonomy faster than their governance frameworks can keep up. Traditional AI governance wasn’t designed for action-level authorization, which is where most agent-specific risk accumulates.Grant Thornton的《2026年AI影响调查》对950名企业高管进行了调查，发现78%的人缺乏强烈信心能在90天内通过独立的AI治理审计。与此同时，企业正以比其治理框架更快的速度提高智能体自主性。传统的AI治理并非为行为级授权而设计，而这正是大多数智能体特定风险积累的地方。

And there’s a hard deadline attached to this. The EU AI Act becomes fully enforceable for high-risk systems in August 2026. Credit scoring, hiring decisions, healthcare support, and critical infrastructure all fall in scope. If your internal platform doesn’t handle conformity assessments, human oversight mechanisms, complete audit trails, and ongoing monitoring, that’s not a v2 feature. That’s a legal exposure.而且这有一个硬性截止日期。欧盟AI法案将于2026年8月对高风险系统全面生效。信用评分、招聘决策、医疗保健支持和关键基础设施都在范围内。如果你的内部平台不处理符合性评估、人工监督机制、完整的审计追踪和持续监控，那不是一个v2功能。那是法律风险。

OWASP now documents “excessive agency” as a top vulnerability class for LLM applications. Cornell researchers have demonstrated indirect prompt injection attacks that manipulate agents through content they ingest. These are agent-specific attack surfaces, and traditional security tooling doesn’t see them.OWASP现在将“过度代理”列为LLM应用程序的顶级漏洞类别。康奈尔大学的研究人员已经演示了间接提示注入攻击，通过智能体摄取的内容操纵它们。这些是智能体特定的攻击面，传统安全工具看不到它们。

RBAC was designed for humans with predictable intent. Agents don’t have predictable intent.RBAC是为具有可预测意图的人类设计的。智能体没有可预测的意图。

Eval评估

The assumption is that evaluation means writing test cases and measuring accuracy. You built software before. You know how to test things.假设是评估意味着编写测试用例和测量准确性。你以前构建过软件。你知道如何测试东西。

Agent evaluation is qualitatively different from traditional software testing or even LLM evaluation, McKinsey’s QuantumBlack team noted: For LLMs, you evaluate the response to a prompt. For a single agent, you evaluate the full trajectory, including tool calls, state transitions, and intermediate decisions. For multi-agent systems, you evaluate system dynamics, including coordination patterns and collective invariants.智能体评估在本质上不同于传统软件测试甚至LLM评估，麦肯锡的QuantumBlack团队指出：对于LLM，你评估对提示的响应。对于单个智能体，你评估完整的轨迹，包括工具调用、状态转换和中间决策。对于多智能体系统，你评估系统动态，包括协调模式和集体不变量。

This matters because agent behavior is nondeterministic by design. The same input produces different valid execution paths. “Did the agent succeed?” is no longer a yes-or-no question, because the agent might reach the right answer through a trajectory you didn’t anticipate, or reach the wrong answer through a trajectory that looks reasonable until the last step.这很重要，因为智能体行为本质上是非确定性的。相同的输入会产生不同的有效执行路径。“智能体成功了吗？”不再是一个是非问题，因为智能体可能通过你未预料到的轨迹达到正确答案，或者通过一个直到最后一步看起来都合理的轨迹达到错误答案。

The tooling ecosystem reflects this. Google Vertex AI has standardized trajectory_exact_match, trajectory_precision, and trajectory_recall as production metrics. These didn’t exist 18 months ago. LangSmith, Braintrust, Arize, Galileo, Maxim, and others are building full evaluation platforms around trajectory-based analysis, LLM-as-judge scoring with statistical validation, and regression testing against production failures.工具生态系统反映了这一点。Google Vertex AI已将trajectory_exact_match、trajectory_precision和trajectory_recall标准化为生产指标。这些在18个月前还不存在。LangSmith、Braintrust、Arize、Galileo、Maxim等公司正在构建完整的评估平台，围绕基于轨迹的分析、带有统计验证的LLM作为评判评分，以及针对生产故障的回归测试。

Here’s the signal that the category is real: LangChain’s 2026 State of AI Agents report found that 57% of organizations now have agents in production, and 32% cite quality as the top deployment barrier. Gartner projects that 60% of software engineering teams will adopt AI evaluation and observability platforms by 2028, up from 18% in 2025. When a category jumps from 18% to 60% adoption in three years, that’s not a “we can build this in a sprint” situation.以下是该类别真实存在的信号：LangChain的《2026年AI智能体现状》报告发现，57%的组织现在在生产中使用智能体，32%将质量列为首要部署障碍。Gartner预测，到2028年，60%的软件工程团队将采用AI评估和可观测性平台，而2025年这一比例为18%。当一个类别在三年内从18%的采用率跃升至60%时，这可不是一个“我们可以在一个冲刺中构建”的情况。

You can’t tell whether your evaluation is working without another evaluation. Judge drift, calibration against human experts, internal consistency across independent runs. . .your eval system needs its own eval system, which is exactly the kind of recursion that eats platform teams alive.你无法判断你的评估是否有效，除非进行另一个评估。评判漂移、与人类专家的校准、独立运行之间的内部一致性……你的评估系统需要自己的评估系统，这正是那种会吞噬平台团队的递归。

Orchestration编排

The orchestration layer hasn’t converged. LangGraph uses directed graphs with conditional edges. CrewAI uses role-based crews. OpenAI’s Agents SDK uses explicit handoffs. AutoGen uses conversational GroupChat. Google ADK uses hierarchical agent trees. Claude’s Agent SDK uses tool-use chains with subagents. Microsoft’s Agent Framework is its own thing. Each represents a different bet on state management, communication pattern, and coordination model. None of them are interchangeable. Migration between them isn’t a config change—it’s rewriting most of your agent logic.编排层尚未收敛。LangGraph使用带有条件边的有向图。CrewAI使用基于角色的团队。OpenAI的Agents SDK使用显式交接。AutoGen使用对话式GroupChat。Google ADK使用分层智能体树。Claude的Agent SDK使用带有子智能体的工具使用链。微软的Agent Framework是它自己的东西。每个都代表了关于状态管理、通信模式和协调模型的不同赌注。它们之间不可互换。在它们之间迁移不是一个配置更改——而是重写你大部分智能体逻辑。

Underneath them, the protocol layer is still being invented. The Model Context Protocol is becoming the standard for tool integration, and agent-to-agent (A2A) protocols are emerging for cross-framework coordination. Both are moving targets, and building on a moving protocol is a cost that internal platform teams rarely price in.在它们之下，协议层仍在被发明。模型上下文协议正在成为工具集成的标准，而智能体到智能体（A2A）协议正在出现，用于跨框架协调。两者都是移动的目标，在移动的协议上构建是内部平台团队很少计入的成本。

If you built your own orchestration layer in 2024, you’re rewriting it in 2026. The teams that picked a framework spent those two years shipping.如果你在2024年构建了自己的编排层，那么你在2026年正在重写它。选择框架的团队在那两年里一直在交付。

The honest case for building诚实的自建理由

I want to engage the strongest version of the build argument, because there are real reasons to build, and pretending otherwise makes this piece less useful than it should be.我想探讨自建论点的最强版本，因为确实有真正的理由去自建，假装不是这样会让这篇文章失去应有的价值。

Proprietary data genuinely is a durable competitive moat. Mastercard built a foundation model on its transaction network. Plaid built one on its financial institution coverage. As Morgan Stanley’s analysis from last year made clear, decades of verified historical data with consistent identifiers is both technically challenging and prohibitively expensive for outside players to recreate. If your organization has data like that, you should absolutely build on it.专有数据确实是一个持久的竞争护城河。万事达卡在其交易网络上构建了一个基础模型。Plaid在其金融机构覆盖范围上构建了一个。正如摩根士丹利去年的分析所表明的，具有一致标识符的数十年的经过验证的历史数据，对于外部参与者来说，在技术上具有挑战性且成本高昂，难以复制。如果你的组织拥有这样的数据，你绝对应该在此基础上进行构建。

Regulated industries have legitimate reasons to want control over the full stack. Off-the-shelf AI tools don’t always cleanly map to frameworks like HIPAA, GxP, 21 CFR Part 11, SOX, FFIEC, and PCI DSS, and the cost of a failed audit is measured in business units shut down, not in sprints.受监管行业有正当理由希望控制整个技术栈。现成的AI工具并不总能清晰地映射到HIPAA、GxP、21 CFR Part 11、SOX、FFIEC和PCI DSS等框架，而审计失败的成本是以业务部门关闭来衡量的，而不是以冲刺来衡量的。

Vendor lock-in at the AI layer is subtler and more dangerous than in traditional software. If your agentic workflows are built on a vendor’s proprietary orchestration layer, switching costs compound rapidly across memory, eval, and integrations simultaneously.AI层的供应商锁定比传统软件更微妙、更危险。如果你的智能体工作流构建在供应商的专有编排层上，切换成本会在记忆、评估和集成方面同时迅速累积。

But here’s the distinction that matters: Those are arguments for building agents on top of platform components, not arguments for building the platform components themselves. You can own the data, the domain logic, the evaluation criteria, the governance policies, and the specific behaviors your business needs without owning the memory layer, the orchestration engine, or the trace collection infrastructure underneath them.但这里有一个重要的区别：这些是在平台组件之上构建智能体的理由，而不是构建平台组件本身的理由。你可以拥有数据、领域逻辑、评估标准、治理策略以及你的业务所需的特定行为，而无需拥有底层的记忆层、编排引擎或追踪收集基础设施。

Build the things that are specific to your business. Buy the things that are specific to the technology category. That’s the heuristic.构建对你的业务特定的东西。购买对技术类别特定的东西。这就是启发式方法。

Five questions before you commit承诺前要问的五个问题

If you’re the platform engineer being pulled into this decision, here are the questions worth asking before anyone signs up for the scope.如果你是正在被拉入这个决策的平台工程师，以下是在任何人签署工作范围之前值得问的问题。

Are you building an agent platform or a workflow system? They’re not the same scope, and conflating them is where most of the cost overruns originate. A workflow system is a reasonable thing to build. An agent platform is four product categories you haven’t staffed for.你是在构建智能体平台还是工作流系统？它们不是相同的工作范围，混淆它们是大多数成本超支的根源。工作流系统是一个合理的东西去构建。智能体平台是你尚未配备人员的四个产品类别。

Can you articulate what “done” looks like for each of the four components? Memory, governance, eval, orchestration. In under three sentences each. If you can’t, you don’t have requirements. You have a vibe. And vibes don’t ship.你能用不到三句话描述每个组件的“完成”是什么样子吗？记忆、治理、评估、编排。如果你不能，你就没有需求。你只有一种感觉。而感觉无法交付。

What happens to your platform when you need to swap the underlying model? Menlo’s December 2025 data shows Anthropic went from 12% of enterprise LLM spend in 2023 to 40% in 2025, while OpenAI fell from 50% to 27%. Enterprises didn’t plan those switches. The capability gaps forced them. If your internal platform hardcoded assumptions about context windows, tool-calling formats, or reasoning styles from one vendor, swapping models isn’t an API key change. It’s simultaneous rewrites across memory, eval, and orchestration.当你需要更换底层模型时，你的平台会发生什么？Menlo的2025年12月数据显示，Anthropic在企业LLM支出中的份额从2023年的12%上升到2025年的40%，而OpenAI从50%下降到27%。企业没有计划这些切换。能力差距迫使了它们。如果你的内部平台硬编码了来自一个供应商的关于上下文窗口、工具调用格式或推理风格的假设，那么切换模型就不是一个API密钥更改。它是在记忆、评估和编排方面的同时重写。

What happens when the techniques themselves change? Eighteen months ago the default pattern was RAG with flat vector retrieval. Now it’s just-in-time context strategies, agent-managed memory tiers, and trajectory-based evaluation. Anthropic’s own follow-up to “Building Effective Agents” explicitly acknowledges the field has moved since they wrote the original. If your platform baked in the 2024 patterns, the 2026 patterns are a refactor, not a config change. Vendor platforms absorb those shifts as releases. Internal platforms absorb them as sprints.当技术本身发生变化时会发生什么？18个月前，默认模式是带有平面向量检索的RAG。现在是即时上下文策略、智能体管理的记忆层级和基于轨迹的评估。Anthropic自己对“构建有效智能体”的后续文章明确承认，该领域自他们撰写原文以来已经发生了变化。如果你的平台固化了2024年的模式，那么2026年的模式是一次重构，而不是配置更改。供应商平台将这些变化作为发布版本来吸收。内部平台将它们作为冲刺来吸收。

What happens when the platform team leaves? This is the tale as old as COBOL, custom ESBs in 2008, or hand-rolled container orchestration in 2015. A small team builds something clever, it works, they move on, and five years later you’re paying premium rates to contractors who can still read the code. Agent platforms are a particularly bad candidate for this pattern because the talent pool is both small and mobile. Here’s the uncomfortable version of the question: Who on your team, today, could rebuild the memory layer if the person who wrote it left tomorrow?当平台团队离开时会发生什么？这是和COBOL、2008年的自定义ESB或2015年的手工容器编排一样古老的故事。一个小团队构建了一些巧妙的东西，它工作了，他们离开了，五年后你支付高价给仍然能读懂代码的承包商。智能体平台是这种模式的特别糟糕的候选者，因为人才库既小又流动。这里有一个不舒服的问题版本：今天，如果你的团队中编写记忆层的人明天离开了，谁能够重建它？

What this looks like in 2 years两年后这会是什么样子

Gartner’s prediction that over 40% of agentic AI projects will be canceled by 2027 isn’t really about the AI. It’s about projects that got scoped before anyone understood the shape of the work. Most of the canceled projects will be internal builds, because internal builds are where the scope estimation error accumulates. Deloitte’s data on two- to four-year AI ROI horizons is the warning shot. If your timeline to value is already long, every month you spend rebuilding a component that exists as a product is a month you don’t have.Gartner预测，到2027年，超过40%的智能体AI项目将被取消，这实际上与AI无关。而是关于在任何人理解工作范围之前就被确定范围的项目。大多数被取消的项目将是内部构建，因为内部构建是范围估算错误累积的地方。德勤关于两到四年AI投资回报率的数据是一个警告信号。如果你的价值实现时间线已经很长，那么你每个月花在重建一个作为产品存在的组件上，就是一个月你无法拥有的时间。

The teams that built their platforms around OpenAI in 2023 weren’t wrong. They made a reasonable bet on the market leader at the time. But they spent 2025 porting to a landscape where Anthropic had tripled share and Google had gone from 7% to 21%. The teams that picked model-agnostic platforms spent 2025 shipping. The only durable bet in this space is the one that assumes the bet will change.那些在2023年围绕OpenAI构建平台的团队并没有错。他们当时对市场领导者做出了合理的赌注。但他们花了2025年迁移到一个Anthropic份额增长三倍、谷歌从7%增长到21%的格局。那些选择模型无关平台的团队花了2025年交付。这个领域唯一持久的赌注是假设赌注会改变的赌注。

The best platform engineering decision you can make this quarter might be to not build the platform.本季度你能做出的最佳平台工程决策可能是不构建平台。

Sources来源

Primary sources主要来源

Secondary Sources次要来源

  • Mem0, “Mem0 Raises $24M to Build the Memory Layer for AI,” October 2025,
    https://mem0.ai/series-a.    Mem0, “Mem0 Raises $24M to Build the Memory Layer for AI,” October 2025, https://mem0.ai/series-a.
  • Felicis, “Felicis’s Seed in Letta,” September 2024,
    https://www.felicis.com/blog/letta.    Felicis, “Felicis’s Seed in Letta,” September 2024, https://www.felicis.com/blog/letta.
  • Vectorize.io, “Mem0 vs Zep,” Benchmark Comparison,
    https://vectorize.io/articles/mem0-vs-zep.    Vectorize.io, “Mem0 vs Zep,” Benchmark Comparison, https://vectorize.io/articles/mem0-vs-zep.
  • Rasmussen et al., “Zep: A Temporal Knowledge Graph Architecture for Agent Memory,” arXiv 2501.13956,
    https://arxiv.org/abs/2501.13956.    Rasmussen et al., “Zep: A Temporal Knowledge Graph Architecture for Agent Memory,” arXiv 2501.13956, https://arxiv.org/abs/2501.13956.
  • OWASP, “LLM08:2025 Excessive Agency,” OWASP Top 10 for LLM Applications,
    https://genai.owasp.org/llmrisk/llm08-excessive-agency/.    OWASP, “LLM08:2025 Excessive Agency,” OWASP Top 10 for LLM Applications, https://genai.owasp.org/llmrisk/llm08-excessive-agency/.
  • Greshake et al., “Not What You’ve Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection,” arXiv 2302.12173, February 2023,
    https://arxiv.org/abs/2302.12173.    Greshake et al., “Not What You’ve Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection,” arXiv 2302.12173, February 2023,https://arxiv.org/abs/2302.12173.
  • Model Context Protocol, Official Specification,
    https://modelcontextprotocol.io.    Model Context Protocol, Official Specification, https://modelcontextprotocol.io.
  • PYMNTS, “FinTechs Race to Build Foundation Models on Proprietary Data,” 2026,
    https://www.pymnts.com/artificial-intelligence-2/2026/fintechs-race-to-build-foundation-models-on-proprietary-data/.    PYMNTS, “FinTechs Race to Build Foundation Models on Proprietary Data,” 2026,https://www.pymnts.com/artificial-intelligence-2/2026/fintechs-race-to-build-foundation-models-on-proprietary-data/.
  • Deloitte, “State of Generative AI in the Enterprise,” Quarterly Reports,
    https://www.deloitte.com/us/en/insights/topics/digital-transformation/state-of-generative-ai-in-enterprise.html.    Deloitte, “State of Generative AI in the Enterprise,” Quarterly Reports, https://www.deloitte.com/us/en/insights/topics/digital-transformation/state-of-generative-ai-in-enterprise.html.
Post topics: AI & ML
译
正在发送到服务端翻译...