汽车工业的下一条流水线，开始由 AI 来安排
1913 年，美国密歇根州的福特 Highland Park 工厂里，工程师们进行了一项看起来有些简陋的实验：他们用绳索和卷扬机牵引汽车底盘，让它缓缓经过不同的装配工位。如此这样，工人们就不再需要围绕静止的汽车来回移动，而是在自己的位置上等待底盘经过，进而完成手头的装配任务。
这个实验起初甚至谈不上精密，但效果十分惊人，经过不断改进，到 1914 年的时候，福特 Model T 车型底盘的装配用工时间，从最初的 12.5 小时缩短到了 93 分钟。
▲ 最早的福特汽车工厂流水线
这就是后来写进了历史书，改变了汽车工业的移动装配流水线。
一个多世纪过去，汽车工业沧海桑田早已不同往日，从内燃机到电机，从机械控制到软件定义汽车，今天的汽车与当初的福特 Model T 已经大相径庭。
但「流水线」的巧思依旧不过时，10 月 29 日，在深圳举行的腾讯全球数字生态大会智慧出行专场上，腾讯发布了全新的智能座舱 TAI 智能体，以及面向汽车行业的 AI 组织效能解决方案，同时公布了智能驾驶、汽车出海、具身智能和 Robotaxi 等领域的最新进展。
腾讯智慧出行横跨 BC 两端，这些业务看起来相距甚远，一端是车主用一句话预订餐厅、规划旅程，另一端则是汽车企业利用 AI 完成研发、生产和经营决策。
▲ 腾讯集团副总裁、腾讯智慧出行总裁钟翔平
不过依旧可以找到一些内在联系：过去需要人在不同软件、部门和系统之间反复沟通的工作，正在被 AI 重新组织。腾讯集团副总裁、腾讯智慧出行总裁钟翔平将其概括为：
产品进化牵引组织变革，组织变革支撑产品进化。
TAI 智能体方案，让汽车学会自己把事情办完
都说有了车之后，生活半径就会扩大许多，假如用户准备在周五晚开车去跨省 4 个小时车程的陌生城市度过一个周末，事情就有趣又复杂了。
在过去，这是一项体力和脑力都耗费不少的劳动密集型的娱乐活动。
我们需要先在社交平台搜索攻略，打开地图研究路线，找餐厅、查营业时间、订酒店、买门票，再把各种地址保存到手机里。
出发之后，还得在导航、音乐、电话和其他服务之间不断切换。即便坐进一辆拥有高性能芯片、几块大屏幕和语音助手的智能汽车，大部分事情依然需要人来完成。
在今年北京车展期间，腾讯已经发布了覆盖点单、向导、娱乐、车主服务等场景的智能体矩阵，尝试把汽车与现实世界的服务连接起来。在这次生态大会上，腾讯又往前走了一步：让智能体从理解需求、调用服务，进一步走向执行任务。
此次升级的 TAI 智能体方案，由 TAI 座舱助手、TAI 出行服务智能体、TAI 内容娱乐智能体，以及跨端接续服务共同组成。
首先升级的是汽车与人交流的能力。
TAI 座舱助手融合了元宝全双工端到端语音、元宝搜索和混元大模型，能够支持低延迟、多音区的自然语言交互，并通过上下文记忆和车端感知，更准确地理解驾驶员的意图。
这里有一个不得不提的重要前提：汽车是一个对安全和注意力有特殊要求的场景。
在发布会前的采访中，腾讯云副总裁、智慧出行负责人钟学丹特别强调：
车内的 AI 需要真正理解用户，同时尽可能减少驾驶过程中的分心。信息来源的准确性和交互效率，比单纯追求聊天的趣味性更重要。
为此，腾讯将元宝搜索引入车端，元宝搜索已接入出行、金融等 13 个垂直行业的 157 个权威信源，试图减少 AI 在真实服务场景中出现的信息错误。
AI 在手机上讲错一个不好笑的冷笑话，顶多让人尴尬；但在驾驶过程中提供错误的道路信息，后果可能严重得多。
交流能力之外，另一项变化发生在任务执行上。
▲腾讯云副总裁李博
腾讯云副总裁李博在发布会上解释了这种任务分工：
腾讯地图负责找得到、腾讯生态负责连接服务、微信支付负责交易闭环。
看起来很简单，实际上链条还挺长，并且这事儿基本上也只有腾讯能做。
具体来说，腾讯地图提供位置搜索、路线规划、周边服务等基础能力，帮助 AI 理解用户的出行需求并找到合适的目的地；微信小程序和第三方服务生态负责连接餐饮、酒店、景区等现实世界的服务，智能体可以进一步完成预订、购票和点单，甚至通过 AI 外呼餐厅完成订座；微信支付则打通交易的最后一个环节。
依旧是用周末跨省游的场景来举例，用户只需要告诉汽车「想在 xx 市找个适合带孩子玩的地方，最好沿途还能吃顿不错的午饭」，AI 就可以结合实时路况、目的地信息、用户偏好和周边服务，推荐合适的景点，规划包含补能和休息的路线，寻找符合口味的餐厅，并协助完成门票购买、餐厅预订和支付。
对用户来说，最理想的情况就是不再需要了解这些功能究竟来自哪个 App、由什么技术实现。「这么近，那么美，周末到河北」或者加个后半句「路好找，店好订，服务全都在腾讯」。
汽车相关的智能体也可以存在于汽车之外。这次腾讯同步升级了 TAI 跨端接续服务，打通手机元宝、微信小微等入口，让用户可以在车外查询车辆状态、传递目的地、同步行程，部分服务在抵达目的地后还可以回到手机上继续使用。
这就意味着，一次旅行未必是从按下汽车启动键开始，也不必在熄火下车时结束。从出门前的准备，到驾驶过程中的服务，再到抵达后的活动，汽车开始成为一套连续服务系统中的组成部分。
钟学丹在采访里提到，未来不同汽车品牌之间的智能化差异，一方面来自车辆自身的感知、车控和硬件能力，另一方面则来自对用户上下文、喜好和习惯的理解。
也就是说，汽车品牌们除了要学会造好车之外，去理解了解用户是同等重要的事情。汽车的智能化价值，开始体现在它能替人完成多少工作，以及让人少花多少精力。
汽车工厂，也需要用 AI 来排出一条新的流水线
车内的智能体正在帮助用户减少操作把体验自动化起来，走进汽车企业内部，AI 面对的则是另一种复杂得多的状况。
比如一个让不少品牌 CEO 丢掉工作的灵魂拷问，下个月应该生产多少辆汽车？
最简单的提问背后，是最复杂的计算，想要回答准确这个问题，需要几乎整个公司都动脑筋。销售部门关心市场订单，生产部门关心产能，采购部门关心供应商交付，财务部门关心库存和现金流。更麻烦的是，各个部门都有自己的数据和工作目标。市场需求随时可能变化，而企业的经营决策必须综合考虑所有因素。
▲ 腾讯云副总裁孙燕
腾讯云副总裁孙燕在采访里提到了一个案例，某汽车大厂原先需要通过大量跨部门会议完成产销协同，产量的决策过程往往要持续两三周，甚至一个月。腾讯与其合作引入智能体后，系统能够综合企业内部经营数据与外部市场信息，围绕利润、销量等不同目标生成决策建议。
这套流程将原先需要数周完成的决策，压缩到了 1—2 天。当然，最终决定仍然由企业管理者作出，AI 主要的作用是承担信息整合、分析和辅助决策的工作。
一家大型汽车企业内部，可能有数万名员工、数百个业务系统，每个人都在努力提高自己的工作效率，但组织的整体效率往往受制于部门之间的信息传递与协作成本。
这种摩擦或许是当下中国汽车工业「卷」过头的一个诱因，个体的效率增加未必促成整体效率的提升。
这也是腾讯此次提出 AI 组织效能解决方案的背景，今年以来，WorkBuddy、CodeBuddy 等 AI 工具已经进入不少汽车企业。腾讯表示，长安、比亚迪、广汽、零跑等近九成国内车企，正在使用相关智能体产品及模型服务。
不仅仅是单个员工能够用 AI 写代码、做表格、整理资料，还需要整个企业真正建立 AI 工作流程，腾讯希望通过企业 Agent 平台，将散落在员工个人手中的 AI 能力进一步连接到企业数据、知识库和业务系统之中。
这里面相当重要的部分是 Skill 的提取和复用，一个经验丰富的工程师可能花费数年积累了一套独特的产品验证方法，即使他愿意分享，新同事也需要经过学习、实践和反复沟通，才能掌握其中的诀窍。
长安汽车代表祁越就提到了一个很现实的情况：
有没有很多知识在专家的脑子里而不在我们的数据库里。
今年 8 月，腾讯与长安汽车共同组建了「AI 火箭班」，通过 FDE 模式让技术人员深入业务场景，围绕产品策划、产销协同、质量运营、AI Coding 等方向，共同开发了八大场景智能体，目前已经上线运营。
腾讯提出的组织效能方法论包含三个层次：员工提效、流程重构和组织转型。腾讯云副总裁孙燕说，判断 AI 有没有真正释放生产力，关键在于业务价值是否实现。
类似的思路，也延伸到了汽车销售环节。此次发布的腾讯企点营销云 MAGIC AI，尝试将用户洞察、内容生产、营销触达和销售线索跟进连接起来。
腾讯披露的首批试点结果显示，相对于对照组，AI 线索策略运营平台帮助门店将线索到店率提升了 24%，单条线索的跟进频次提升了 53%。
从研发、生产到营销，汽车企业内部原先彼此分散的工作，开始出现新的连接方式，这让人想到前面说的福特流水线。
100 多年前，福特通过重新安排工序，让零件与工人以更高效的方式相遇。今天，AI 则开始尝试让分散的数据、知识与业务需求，更快速地找到彼此。
有所不同的是，工业流水线追求的是高度确定性的重复操作，今天的 AI 要面对不断变化的市场、目标和业务环境。它需要处理的复杂程度远远超出了当年那根牵引底盘的绳索……
从司机乘客，到汽车工厂，再到整个产业，AI 还可以走多远？
把视野从一辆汽车的起点、到一家车企角度继续放大，其实会发现腾讯此次发布的很多内容，都在围绕一个问题展开：中国汽车市场进入了更加激烈的存量竞争阶段，新的增长空间在哪里？
腾讯云副总裁李博在大会上给出了三个方向：国内市场的差异化体验、汽车企业的全球化，以及具身智能和 Robotaxi 等新场景。
过去几年，汽车行业的智能驾驶技术路线经历了从 BEV、端到端到 VLA、世界模型的快速迭代。模型能力越来越强，随之而来的是更庞大的训练需求、更复杂的数据闭环，以及持续增长的算力成本。
腾讯披露，其智能驾驶云业务在 2025 年的规模增长超过 300%。通过全链路智算解决方案，腾讯帮助一家国际头部汽车零部件企业将混合算力集群利用率提高到 98.4%，推理场景 GPU 效率提升超过 60%。
节省的成本，最终还是会和消费者息息相关。
出海市场也有类似的需求。
如今，中国汽车企业已经能够把成熟的电动车产品销售到世界各地，同时，海外用户需要本地化的地图、内容和服务，智能辅助驾驶研发也需要应对不同国家的数据法规、基础设施和运营环境。
中国市场里已经成熟的智能化体验，到了海外未必能够直接复制。为此，腾讯此次升级了海外智能化解决方案，覆盖海外智驾云、营销服务、地图和效率智能体等方向。
其中，海外智驾云提供数据回传、处理、训练、仿真及合规工具链；腾讯智驾地图海外版则计划在 2026 年底推出首个版本，试图降低车企在不同市场进行智能化适配的成本。
腾讯还将部分 AI 能力延伸到具身智能与 Robotaxi。
在具身智能领域，腾讯推出了数据闭环套件，并结合 Tairos 具身智能开放平台，为机器人研发提供数据处理、生成与训练支持。在 Robotaxi 领域，腾讯则通过云、地图和出行服务平台连接自动驾驶企业与用户，小马智行、文远知行等企业已经接入腾讯出行平台。
汽车、机器人和 Robotaxi 所面对的技术问题并不完全相同，但它们确实拥有一些共同的基础能力：感知真实世界、理解环境、作出决策，并在物理空间中执行任务。特别是在模型训练、仿真、数据处理和算力管理这些环节，汽车产业多年积累的能力，有机会继续向更多场景延伸。
钟翔平在大会上说：
车企负责造好车，我们负责让 AI 用得起来、用出效果。
腾讯不需要亲自制造汽车，也不打算自己运营一支 Robotaxi 车队。它希望提供的是连接不同汽车企业、技术能力与服务生态的基础设施。
一百多年前，福特让底盘沿着生产线前进，改变了汽车的制造方式。今天，AI 开始让数据、知识和服务围绕真实需求流动。
汽车工业的下一条流水线，已经不再需要一根看得见的绳索。
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。
The next water line in the auto industry is set by AI.
In 1913, at the Ford Highland Park factory in Michigan, United States, engineers carried out an experiment that seemed to be a little simplistic: They used ropes and rollers to pull the chassis of the car to slow it through different assembly positions. In this way, the workers will no longer have to move around static cars, but will wait in their own positions for the chassis to pass, thus completing the assembly tasks at hand.
The experiment was not even sophisticated at first, but it was a remarkable improvement, and by 1914 the fit time for the Ford Model T chassis had been reduced from the first 12.5 hours to 93 minutes.
The first Ford Autowork line.
This was later written into a history book that changed the mobile assembly and flow line of the automobile industry.
More than a century ago, the automobile industry was different, from the internal combustion engine to the electric engine, from mechanical control to software defining the car, and today's car is quite different from Ford Model T.
On October 29, at the Tencent Global Digital Ecology Congress in Shenzhen, the Tencentpublish launched a new smart cabin, TAI AI agent, and an AI organizational effectiveness solution for the automobile industry, as well as the latest developments in the areas of smart driving, car sailing, smartness, and Robotaxi.
Tencent intelligently travels across both ends of the BC, and these operations appear to be far apart, with the owner booking the restaurant in one sentence, planning the journey, and the car company using AI to make R & D, production and operating decisions.
Mr. Zhong Zhengping, Vice President of Tencent Group, and President of Tencent's Smart Travel
However, some intrinsic links can still be found: the work that used to require repeated communication between different software, departments and systems is being reorganized by AI. The Vice President of the Tencent Group and President of the Tencent Smart Traveling Group summarized it as follows:
Product evolution leads to organizational change, which supports product evolution.
TAI AI agent, let the Automotive Society finish the job.
It's said that with a car, the life radius will expand a lot, and it's interesting and complicated if users are going to drive across the province for a weekend in a strange city of four hours on Friday night.
In the past, this was a labour-intensive recreational activity with considerable physical and intellectual costs.
We need to search social platforms, open maps, find restaurants, check hours, book hotels, buy tickets, and keep addresses on the phone.
After departure, there is also a constant switch between navigation, music, telephone and other services. Even in a smart car with high-performance chips, large screens and voice assistants, most things still need to be done.
During this year ' s Beijing Car Fair, Tencent had publish as a matrix of AI agent scenarios covering points, guides, entertainment, car owner services and so on, trying to connect cars to the real world ' s services. At this ecological conference, Tencent took another step forward: to move AI agent further from understanding needs, calling services and implementing mandates.
This upgrade of the TAI AI agent programme consists of TAI cabin assistants, TAI travel service AI agent, TAI content entertainment AI agent, and cross-end succession services.
The first upgrade is the ability of cars to communicate with people.
The TAI cabin assistant, which combines the full binary-to-end voice, the binary search and the hybrid LLM, is able to support the natural linguistic interaction of the low-lated, multisound area and to understand the driver ' s intentions more accurately through context memory and car-side perceptions.
There is an important premise to mention here: the car is a scene with special requirements for safety and attention.
In an interview before publish, Vice-President Tencent Yun and Head of the Smart Travels, Zhong Shuen, highlighted in particular:
The AI in the car needs a real understanding of the user while minimizing the distraction during the drive. The accuracy and interactive efficiency of the information sources are more important than simply pursuing the fun of chatting.
To that end, Tencent introduced the betel search to the end of the car, where it was accessed by 157 authoritative sources in 13 vertical industries, such as travel, finance, etc., in an attempt to reduce AI ' s information errors in the real service scene.
AI misspokes a non-viable cold joke on the phone, which is embarrassing at best; but providing the wrong road information during the drive may have far more serious consequences.
In addition to communication capacity, another change occurred in mandate implementation.
Vice President Li Bo, Tencent Cloud
Vice President Tencent Lee Bo explained this division of tasks at publish:
Tencent maps are available, Tencent ecology is responsible for connection services, and WeChat is responsible for trade closure.
It looks simple, actually, but the chain is quite long, and it's basically Tencent.
Specifically, the Tencent map provides basic capabilities such as location search, route planning, perimeter services, etc., to help AI understand the user ' s travel needs and find a suitable destination; WeChat microprograms and third-party service ecology are responsible for linking services to the real world such as catering, hotels, landscapes, etc., AI agent can further complete bookings, ticketing and billing, or even bookings through the AI Call Dining Room; WeChat pays for the final part of the transaction.
It is still an example of a weekend trip across the province, where a user simply tells the car that "in order to find a place in xx that is suitable for children, and preferably have a good lunch along the way", AI can combine real-time road conditions, destination information, user preferences and peripheral services, recommend suitable attractions, plan routes that include supplementation and rest, search for delicious restaurants and assist in the purchase, reservation and payment of tickets.
For users, it would be preferable to know no longer which app and which technology these functions are derived from. "So close, so beautiful, go to Hebei on weekends" or add the second sentence, "Road, shop, all services are in Tencent."
Car-related AI agent may also exist outside the vehicle. This time, Tencent synchronized the upgrade of the TAI cross-end relay service, which makes it possible for users to check the vehicle ' s status outside the vehicle, to transmit the destination, to synchronize the journey, and to return to the mobile phone after reaching the destination.
This means that a trip does not necessarily start with pressing the car start button or end when the engine goes out. From pre-departure preparation to driving services, to post-arrival activities, cars became part of a continuous service system.
In his interview, Zhong Xue Dan mentioned that the future intellectual differences between different car brands arise from the vehicle ' s own perception, control and hardware capabilities, on the one hand, and the understanding of user context, preferences and habits, on the other.
In other words, in addition to learning how to build a car, it is equally important to understand users. The intelligent value of a car begins to be reflected in the amount of work it can do for people and the amount of effort it can spend on people.
A car factory, too, needs an AI to drain a new water line.
AI agent in the car is helping users reduce their operations and automate their experiences into the car business, and AI is facing a much more complex situation.
How many cars should be produced next month?
Behind the simplest question is the most complex calculation, and to answer the exact question requires the thinking of almost the whole company. Marketing is concerned with market orders, production is concerned with capacity, procurement is concerned with supplier delivery and finance is concerned with inventory and cash flows. Even more problematic is the fact that each sector has its own data and work objectives. Market demand can change at any time, and business decisions must take all factors into account.
Vice President Sun Yan, Tencent Cloud
In an interview with Vice-President Tencent. Sun Yan mentioned a case where a major automobile factory had previously had to complete production and distribution synergies through a large number of cross-sectoral meetings, and production decision-making often lasted two or three weeks, or even one month. The introduction of Tencent in cooperation with AI agent enabled the system to integrate business data within the enterprise and external market information, generating decision-making recommendations around different objectives, such as profit, sales, etc.
The process compresses the decision-making process, which originally took weeks to complete, to 1-2 days. Of course, the final decisions are still made by business managers, and AI ' s main role is to undertake information integration, analysis and supporting decision-making.
Within a large automobile enterprise, there may be tens of thousands of employees and hundreds of business systems, and everyone is trying to improve their own efficiency, but the overall efficiency of the organization is often constrained by the cost of inter-departmental communication and collaboration.
This friction may be a contributing factor to the current “rolling” of the Chinese automobile industry, and an increase in individual efficiency does not necessarily lead to overall efficiency gains.
This is also the context in which Tencent has proposed an AI organizational effectiveness solution, and since this year, the WorldBuddy, CodeBuddy and other AI tools have entered a number of automobile businesses. Tencent states that nearly 90 per cent of domestic vehicles, such as Changan, Biadi, Guangya, and Zero Run, are using AI agent products and modelling services.
Not only is it possible for individual employees to write codes, form and organize information using AI, but it is also necessary for the entire enterprise to actually establish AI workflow, which Tencent wants to further connect the AI capabilities scattered through the Enterprise Agent platform to the data, knowledge base and business systems of the enterprise.
A significant part of this is the extraction and reuse of Skill, and it may take several years for an experienced engineer to accumulate a unique product validation methodology that, even if he is willing to share it, new colleagues need to learn, practice and communicate repeatedly in order to master the know-how.
The more the Changan Automobile points to a reality:
Is there a lot of knowledge in the minds of experts and not in our databases?
In August of this year, Tencent, together with Changan Motors, set up an "AI Rocket Course" to bring technicians into the field of operations through the FDE model, developing eight major scenarios, which are now operational online, around product planning, marketing synergies, quality operations, and AI Coding.
The approach to organizational effectiveness proposed by Tencent consists of three levels: staff effectiveness, process re-engineering and organizational transformation. Vice President Tencent Sun Yan said that the key to judging whether AI is really releasing productivity is whether business values are being realized.
Similar thinking extends to the car sales chain. This time publish, the Tencent Enterprise Marketing Cloud MAGIC AI, attempts to link user insight, content production, marketing contacts and sales trails.
The first pilot results disclosed by Tencent show that the AI Thread Strategy Operating Platform helped the door shop to increase the number of leads to shop by 24 per cent and the frequency of follow-up of individual leads by 53 per cent compared to the control group.
New ways of connecting, from research and development to production to marketing, have begun to emerge from previously dispersed work within automobile enterprises, which reminds us of the Ford Stream line described earlier.
More than 100 years ago, Ford met with workers in a more efficient manner by reordering the process. Today, AI is beginning to try to find each other more quickly with dispersed data, knowledge and business needs.
The difference is that industrial currents pursue highly determinative duplicate operations, and today AI faces changing markets, targets and operating environments. It's a lot more complex than the one that took the chassis...
How far can AI go from the driver's passenger to the car factory to the whole industry?
By widening the horizon from the point of departure of a car to the point of view of a car company, much of Tencent's content of this publish is about the question: where is the new growth space when the Chinese automobile market enters a more competitive stock phase?
Vice President Tencent Lee Bo gave three directions at the Congress: the experience of domestic market differentiation, the globalization of automobile companies, and new scenarios such as smart and Robotaxi.
Over the past few years, the automobile industry has experienced a rapid trajectories of intelligent driving techniques from the BEV, end-to-end to VLA, world models. The growing capacity of models is accompanied by larger training needs, more complex data closed loops, and the computing costs of sustained growth.
Tencent reveals that its intelligent cloud-driving business grew by more than 300 per cent in 2025. Tencent helped an international head car spares enterprise to increase the utilization of hybrid clusters to 98.4 per cent and the reasoning scene GPU efficiency to over 60 per cent.
The cost savings are ultimately linked to consumers.
There is similar demand in the offshore market.
Today, Chinese automobile companies have been able to sell mature electric car products around the world, while overseas users need localized maps, content and services, and smart-assisted driving research and development needs to respond to data regulations, infrastructure and operating environments in different countries.
The mature intellectual experience in the Chinese market may not be directly replicated abroad. To this end, Tencent has upgraded its overseas smart solutions to cover such directions as intellectual driving abroad, marketing services, maps and efficiency AI agent.
Among them, the overseas smart driving cloud provides data retrieval, processing, training, simulation and compliance tool chains; and the Tencent smart driving map overseas is scheduled to be launched in its first version by the end of 2026, in an attempt to reduce the cost of intelligent matching by car operators in different markets.
Tencent also extends some of the AI capabilities to intelligence and Robotaxi.
In the area of smartness, Tencent has launched a data closed loop suite and, in conjunction with the Tairos smart open platform, provides data processing, generation and training support for robotics development. In the area of Robotaxi, Tencent connects auto-driving enterprises with users via cloud, map and travel service platforms, and companies such as the Ma's Smart and Man's Faraway Departments have access to Tencent travel platforms.
The technical problems faced by automobiles, robots and Robotaxi are not identical, but they do have some common foundational capabilities: understanding the real world, understanding the environment, making decisions and carrying out tasks in physical space. Particularly in the areas of modelling training, simulation, data processing and computing, the ability of the automobile industry to accumulate over the years has the opportunity to continue to extend to more scenes.
Zhong Zhouhei said to the General Assembly:
Car companies build cars, we use AI, we use it.
Tencent did not have to make the car himself, nor did it intend to operate a convoy of Robotaxi itself. It wishes to provide infrastructure linking different automobile enterprises, technological capabilities and service ecology.
More than a hundred years ago, Ford moved the chassis along the production line, changing the way cars were manufactured. Today, AI begins to move data, knowledge and services around real needs.
The next water line in the automobile industry no longer requires a visible rope.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.