Introducing AIDC ResearchAIDC Research 上线
How AI is built, deployed and used inside real organisations. Every paper ships with its data and code.研究 AI 在真实组织里怎样被建造、部署和使用。每篇论文都附上数据和代码。
September 30, 2026 · 2 min read2026 年 9 月 30 日 · 阅读约 2 分钟
Today we are launching AIDC Research.
We study how AI is built, deployed and used inside real organisations: the infrastructure that lets agents persist and scale, how organisations of agents govern and improve themselves, and which kinds of work, in which industries, AI actually takes on. Every paper ships with its data and the code that produces every number in it.
Four areas
- Economics of AI: where AI is used across the economy, what the work it touches is worth, and which kind of AI each industry needs.
- Agent systems: runtimes, orchestration and evaluation for agents that persist, scale and work together.
- Organizations: how organisations of agents share tools, improve themselves and stay governed.
- Domain intelligence: bringing agents to specialist domains without retraining, starting with life cycle assessment.
The first papers
The launch includes seven papers and surveys. Among them:
- Precision or Automation? The Value–Population Ratio and the Demand for AI across Industries, a study of 65 US private industries and how much AI each uses per worker.
- AIDH: An Agent-Independent Distributed Harness for Persistent Agents, on keeping no instance state between requests so that a persistent agent costs in proportion to its work.
- Recursive Self-Improvement in Self-Organizing Multi-Agent Systems, on organisations of agents that package what they build as governed, tested and reversible components.
- From Language Models to Agent Systems, a survey of 434 works across training, architectures, orchestration and open problems.
Why a deployment company does research
Deploying AI raises questions that product work alone cannot settle: which work is worth automating, how agents should be organised, what it costs to keep them running. We would rather answer those questions in the open, with data anyone can check, than with claims nobody can.
All papers are on the Research page.
今天，AIDC Research 上线。
我们研究 AI 在真实组织里怎样被建造、部署和使用：让智能体持续存在、规模化运行的基础设施；智能体组织怎样治理和自我改进；以及在哪些行业、哪类工作上，AI 真正接得住。每篇论文都附上数据，和算出其中每一个数字的代码。
四个方向
- AI 经济学： AI 在经济中用在哪里，它触及的工作值多少，每个行业需要哪一类 AI。
- 智能体系统： 让智能体持续存在、规模化、协同工作的运行时、编排与评估。
- 组织： 智能体组织怎样共享工具、自我改进、保持受治理。
- 领域智能： 不重新训练，把智能体带进专业领域，从生命周期评价开始。
第一批论文
首批共七篇论文与综述，其中包括：
- 《Precision or Automation? The Value–Population Ratio and the Demand for AI across Industries》：研究美国 65 个私营行业，以及每个行业人均用多少 AI。
- 《AIDH: An Agent-Independent Distributed Harness for Persistent Agents》：请求之间不保留实例状态，让一个持续存在的智能体，成本只与它的工作量成正比。
- 《Recursive Self-Improvement in Self-Organizing Multi-Agent Systems》：智能体组织把自己造出的能力打包成受治理、经过测试、可撤回的组件。
- 《From Language Models to Agent Systems》：一篇覆盖 434 项工作的综述，涵盖训练、架构、编排和开放问题。
一家部署公司为什么做研究
部署 AI 会遇到一些单靠做产品回答不了的问题：哪些工作值得自动化，智能体该怎么组织，让它们一直运行要花多少钱。我们宁愿公开回答这些问题，附上谁都能检查的数据，而不是给出谁都没法检查的说法。
全部论文见研究页面。论文为英文。
AIDC · AI Deployment CompanyAIDC · AI Deployment Company
