10 月 22 日，由 InfoQ 极客传媒主办，联合中欧国际工商学院 EMBA 课程部、独角兽创业营、TGO 鲲鹏会共同发起的 InfoQ AI 侦察团 @QCon 上海 2026 首次集结。
从 QCon 上海主论坛，到午餐圆桌、技术专场和闭门私董会，成员们在一天中听技术、聊实践、谈问题，也尝试从其他同侪的经验中寻找企业 AI 落地的解决思路。
AI 侦察团的每个环节各有亮点，关注的却始终是同一件事：当 AI 技术不断发展，企业该如何结合自身业务，把 AI 的技术能力转化为实际的企业价值？
一、从技术趋势中寻找启发
上午，侦察团成员走进 QCon 上海 2026 主论坛，与 1200 余位 AI 领域从业者齐聚一堂，聆听来自阿里巴巴、涛思数据、小影科技、上海交通大学人工智能学院等头部科技企业与高校的技术专家分享 AI 前沿进展与实践经验。
从模型和 Agent 能力的演进，到数据、系统与业务流程的协同，主论坛提供了观察 AI 技术发展的多个视角。对于企业而言，技术能力的提升是 AI 落地的起点，但企业还需要回答：具体选择哪些应用场景？需要哪些基础能力？以及如何与现有业务结合？
技术趋势提供方向，真正落地还要回到企业的实际需求中。
二、把 AI 落地的真问题摆到桌面上
午餐会上，不同背景的企业决策者和 AI 实践者围桌交流，将话题从技术趋势拉回企业自身。
成员们从各自的岗位、企业 AI 应用进展和当前遇到的问题谈起，很快就延伸到更具体的业务问题：哪些场景值得优先尝试？如何推动员工改变工作方式？哪些流程适合引入自动化？投入之后，又该如何衡量实际效果？
不同企业所处的阶段不同，遇到的难题也各有差异。面对面的交流，让成员们有机会对照彼此的实践，重新审视自己的问题，也为下午的技术专场带来了更明确的关注点：其他企业是怎么做的？哪些经验值得借鉴？哪些方法可以带回去进一步验证？
三、四场分享，拆解 AI 落地实践
下午，为 AI 侦察团特别设计的技术专场正式开启。以“帮你把 AI 这道题从头做一遍”为主题，来自阿里云、亚马逊云科技、去哪儿、腾讯及优码云的一线实战专家，带来了四场精彩分享。现场座无虚席，观众参与热情高涨。
1. 让 Agent 成为企业生产力
阿里云智能集团高级解决方案架构师崔飞飞（虚缇）分享了 Agent 在企业业务场景中的实践。他表示，对于企业来说，关注点不应只停留在 Agent 能完成什么任务，还要看它如何结合业务知识、调用相关工具、融入岗位流程，以及能否带来实际效果。
2. 个人提效、团队复利：亚马逊的 AI 实践分享
亚马逊云科技生成式 AI 解决方案架构师高云怡（ Gao Yunyi ）分享了亚马逊在 AI 应用与组织协作方面的经验。她指出，个人提效是起点，让经验沉淀、方法复用，并进一步转化为团队能力，才有机会形成真正的组织复利。
3. AI 在研发组织落地的三段式路径：从人使用 AI 干活到 AI 自主交付
去哪儿基础架构负责人、技术总监李佳奇围绕研发组织的 AI 实践展开分享，探讨 AI 如何从辅助开发者完成具体任务，逐步参与更完整的研发流程与交付环节。
4. AI 驾驭之道：告别单点外挂，重塑组织系统
腾讯 WorkBuddy Nova 大使彭超与优码云 CTO、创始人韦优带来联合分享，讨论企业如何跳出单点工具应用的局限，重新思考 AI 与业务流程、团队协作及组织机制之间的关系。
四场分享从 Agent 应用、团队提效、研发交付和组织协作等角度切入，呈现了企业探索 AI 落地的不同路径。每场分享结束后，现场观众踊跃提问，围绕实践中的具体问题与嘉宾展开交流。
对企业而言，使用 AI 工具只是开始。随着 AI 进入更多业务流程，如何优化协作方式、积累可复用的经验，才是进一步释放价值的关键。
*技术专场更多内容，敬请关注 QCon 上海官网：_https://qcon.infoq.cn/2026/shanghai/track/2012
四、成员访谈，把现场收获带回企业
技术专场结束后，InfoQ 与四位侦察团成员进一步交流。有成员希望打通分散在不同平台的数据，也有成员正在思考如何把个人提效推广到整个组织。此外，客户知识库建设、研发流程优化和职能工作自动化，同样是他们关注的重点。四位成员所处行业和企业 AI 应用阶段不同，但他们都在探索同一个问题：如何找到适合企业的切入点，让 AI 真正解决业务难题。
对 AI 侦察团而言，现场交流带来的启发，最重要的价值是能找到带回企业验证的下一步。
更多内容，详见《_InfoQ AI 侦察团现场手记：工具之外，企业 AI 落地还要跨过哪些关？》
五、聚焦真实难题，共同寻找解法
晚间，InfoQ AI 侦察团成员齐聚闭门私董会，围绕“战略与投入”“业务与 ROI”“组织与人”“规模化与治理”四个方向，讨论企业在 AI 转型中遇到的实际问题。
成员们先提出各自正在面对的难题，再通过投票选出最值得深入讨论的问题。围绕具体案例，大家进一步梳理问题背景、分析成因与阻力，并结合各自的经验提出建议，寻找可以尝试的行动方向。
一位参与者在会后表示，自己所在小组围绕一个真实案例展开的讨论“非常精彩”，大家一起出主意、想办法，让他“受益匪浅”。
私董会的价值正在于此：现场不急于给出标准答案，而是借助不同领域的经验，帮助问题提出者打开思路，找到接下来值得继续探索的方向。
六、带着新思路，持续探索企业 AI 落地
从主论坛的技术分享，到午餐圆桌的经验交流；从下午高密度的案例分享，到晚间围绕真实问题的闭门讨论，InfoQ AI 侦察团 @QCon 上海 2026 将技术、业务与组织层面的讨论连接在一起。
对企业决策者而言，企业 AI 转型没有一套标准答案，更重要的是看清自身所处的 AI 转型阶段，找到值得优先解决的问题，再通过实践判断哪些方法真正有效。这也是 InfoQ AI 侦察团的意义所在：帮助成员们更清楚地识别各自企业的问题，把交流中的启发转化为下一步行动。
未来，InfoQ AI 侦察团将持续连接技术内容、具体实践与企业决策者，让真实经验彼此启发，让 AI 探索更贴近业务，助力企业在 AI 时代长足发展。
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
On 22 October, the InfoQ AI Reconnaissance Group @QCon Shanghai 2026, co-sponsored by the EMBA Curriculum Department of the Central European Institute of International Business and Industry, the Unicorn Enterprise and the TGO Association, gathered for the first time.
From the QCon Shanghai Master Forum to the lunch table, the technical venue and the closed private board, members listen to technology, talk about practice, talk about issues in one day, and try to find solutions for the fall of the business AI from the experience of other peers.
AI ' s every link of the Reconnaissance Corps is bright, but it's always the same thing: When AI technology is evolving, how can an enterprise combine its own business and translate AI ' s technical capabilities into real corporate value?
I. Looking for inspiration from technological trends
In the morning, members of the Recon Corps entered the QCon Shanghai 2026 main forum, where they met with more than 1,200 AI field practitioners to listen to the first technology companies from Alibaba, Toth data, small film technology, Shanghai University School of Artificial Intelligence and technical experts from higher education to share the advances and practical experiences of AI.
From the evolution of models and Agent capabilities to the synergy of data, systems and business processes, the main forum provides multiple perspectives for observing AI technology development. For an enterprise, the upgrading of technological capabilities is the starting point for AI to land, but an enterprise needs to answer: which application scenarios are specifically selected? What basic capacities are needed? and how to integrate with existing operations?
Technological trends provide direction and a return to the real needs of enterprises.
II. Putting the real problem of AI landing on the table
At lunchtimes, business decision makers and AI practitioners from different backgrounds meet around the table to bring the topic back from technological trends to the enterprise itself.
From the point of view of their respective positions, the progress of their corporate AI application and the problems currently encountered, members quickly extended to more specific operational issues: which scenarios should be given priority? How can staff change their ways of working? What processes are suitable for introducing automation? How can actual results be measured after input?
Different enterprises are at different stages and the challenges encountered vary. Face-to-face interaction, which gives members the opportunity to revisit their problems against each other ' s practices, also brings a clearer focus to the afternoon technical session: how do other businesses do they do it? What lessons are worthy of learning? What methods can be brought back for further validation?
3-4 sharing, dismantling AI landing practice
In the afternoon, the technical specialty designed for the AI Reconnaissance Mission was officially opened. With the theme "Help you get the AI from the top," first-line field experts from Alibaba Cloud, Amazonian Cloud Technologies, where to go, Tencent and the Magnetic Clouds brought four wonderful stories. The floor was open and the audience was enthusiastic.
1. Making Agent an enterprise producer Power
The Alibaba Cloud Smart Group Advanced Solutions Architect, Choi Fei, shared Agent's experience in the business landscape. He said that for enterprises, the focus should not only be on what Agent could do, but also on how it combined business knowledge, access to relevant tools, integration into job processes, and the practical impact.
2. Personal effectiveness, team recovery: AI sharing of practices in the Amazon
Gao Yunyi shared Amazon experience in AI applications and organization collaboration. She noted that personal effectiveness was the starting point, allowing experience to sink, re-utilise and further transform into team capacity, thus providing an opportunity for real organizational resilience.
3. AI Three-part path to R & D organization landing: from people using AI to AI self-delivery
Where is Li Jiaqi, Director of Infrastructure and Technology, sharing the AI practice of R & D organizations on how AI can accomplish specific tasks from supporting developers and gradually engage in more complete R & D processes and delivery?
4. AI Managing the Way: Dismissals, reshaping organizational systems
Tencent WorkBuddy Nova, Ambassador Peng Chao and the Macro cloud CTO, founder of Wei Weijo, brought together to discuss the limitations of how companies could jump out of the application of single-point tools and rethink the relationship between AI and business processes, teamwork and organizational mechanisms.
Four sharing sessions, from the perspective of Agent applications, team effectiveness, R & D delivery and organizational collaboration, presented different paths for firms to explore AI landings. At the end of each of the sharing sessions, there was a live audience to question and interact with guests around specific issues in practice.
For an enterprise, using an AI tool is only the beginning. As AI enters more business processes, how to optimize collaboration and build replicable experience is key to further unlocking value.
*For more information from the technical expertise, please pay attention to QCon Shanghai Official Network: https://qcon.infoq.cn/2026/shanghai/track/2012
IV. PARTICIPATION OF MEMBERS TO TAKE THE INSTITUTIONAL ACCOUNT BACK TO BUSINESS
Following the technical expertise, InfoQ further interacted with four members of the reconnaissance mission. Some members wanted to access data scattered across different platforms, while others were thinking about how to extend individual effectiveness across the organization. In addition, the development of a client knowledge base, the optimization of research and development processes and the automation of functional work are also central to their attention. Four members are in different industries and enterprises at different stages of application, but they're all exploring the same question: how to find an entry point suitable for an enterprise and get AI to really solve business problems.
For AI Reconnaissance, the most important value of the inspiration provided by on-site communication is the availability of the next step that was brought back to the enterprise for validation.
For more details, see InfoQ AI: Beyond the tools, what do enterprises AI cross?
V. Focusing on real challenges and finding solutions together
In the evening, the members of the InfoQ AI Reconnaissance Corps gathered together to discuss the practical problems faced by enterprises in the AI transition in the four directions of “Strategy and input” “Business and ROI” “Organization and people” “Sizing and governance”.
Members first raise the challenges that they are facing and then choose the issues that merit the most in-depth discussion by voting. On a case-by-case basis, the context of the problem, the causes and resistance were further refined, and recommendations were made in the light of their respective experiences to find possible courses of action.
Following the meeting, one participant stated that the discussions in his team around a real case had been “excellent” and that the group had come together to come up with ideas and ways to “beneficiary” him.
This is where the value of the private board is: rather than rushing to provide standard answers, it is using experience in different fields to help askers to open their minds and find a way forward that is worth exploring.
VI. CONTINUES TO EXPERIENCE IN ENTERPRISE AI
From technical sharing at the main forum to the exchange of experiences at the lunch table; from high-density case sharing in the afternoon to closed discussions around real issues at night, InfoQ AI Reconnaissance @QCon Shanghai 2026 links discussions at the technical, operational and organizational levels.
For business policymakers, the AI transition does not have a standard set of answers, and it is more important to see the AI transition phase in which they are in, to find priority problems, and to judge by practice which methods are really effective. This is also the point of the InfoQ AI Reconnaissance Mission: to help members identify more clearly the problems of their respective enterprises and to turn the inspiration of exchanges into action next step.
In the future, the InfoQ AI Reconnaissance Mission will continue to connect technical content, concrete practices and business decision makers, so that real experience can inspire one another, and AI can explore closer business and help enable firms to develop in the AI era.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.