本文为原文前 6,000 字符的节选翻译，完整内容请查看原文。

GM is betting compute and AI will help it compete with Chinese automakers

When GM announced its new fourth-generation EV platform on Wednesday, what the automaker didn’t share was almost as notable as what it did. The new EV platform, part of a series of technology-focused announced during its annual GM Forward event in New York, didn’t talk about suspension geometries, dash-to-axle ratios, and torsional rigidity, all nerdy automotive terms that translate to how people will experience a car. Is it supple or sporty? Does it have a refined stance? When it hits a pothole, does it feel like a bank vault or a paper plate?

通用汽车周三发布其第四代电动汽车平台时，这家汽车制造商未透露的信息与已发布的信息同样引人注目。作为在纽约举行的年度“GM Forward”活动中一系列技术发布的一部分，该平台并未提及悬挂几何结构、仪表板到车轴的比例以及扭转刚度等专业术语，这些术语通常决定了人们对汽车的驾驶体验。它是柔软还是运动？姿态是否优雅？当压过坑洼时，感觉像银行金库般稳固还是像纸盘般脆弱？

Instead, GM’s announcement centered on novel battery chemistries, revamped electric motors, and an entirely new central computing platform and electrical architecture that will connect every system in the car, including acceleration and power, steering, braking, and active safety features as well as software. Lots and lots of software. “We think vehicles can become a more capable and helpful part of everyday life if we make them spaces that understand context, reduce friction, and give people back moments that would otherwise be consumed by the demands of the road,” Sterling Anderson, executive vice president, global product, and chief product officer at GM, told TechCrunch in a recent interview.

相反，通用汽车的公告重点在于新型电池化学成分、改进的电动机，以及一个全新的中央计算平台和电气架构。该架构将连接车内的所有系统，包括加速、动力、转向、制动、主动安全功能以及大量的软件。通用汽车全球产品执行副总裁兼首席产品官斯特林·安德森（Sterling Anderson）在最近接受 TechCrunch 采访时表示：“我们认为，如果我们能让车辆成为理解环境、减少摩擦并为人们节省原本会被道路需求所消耗的时间的空间，它们就能成为日常生活中更有能力、更有帮助的一部分。”

That sure doesn’t sound like a traditional automaker’s pitch. Then again, since the arrival of Tesla, Rivian, and Chinese automakers, the industry hasn’t been the same. Ride and handling still matter, but they’re table stakes. What sets vehicles apart today is largely defined by software. Anderson has talked about this future GM before. Last year, the company announced plans to overhaul the electrical and computational guts of its future vehicles in a bid to deliver faster software, more capable automated driving features, and a custom, conversational AI assistant.

这听起来确实不像传统汽车制造商的推销辞令。不过，自特斯拉、Rivian 和中国汽车制造商崛起以来，整个行业已今非昔比。乘坐舒适性和操控性固然重要，但它们只是入场券。如今，决定车辆差异的关键主要在于软件。安德森此前曾谈到过通用汽车的这一未来愿景。去年，该公司宣布计划彻底改造未来车辆的电气和计算核心，旨在提供更快的软件、更强大的自动驾驶功能以及定制的对话式人工智能助手。

This year, GM is providing more details and dates. This technological overhaul is set to debut in 2028 with the Cadillac Escalade IQ. The company will also rollout an autonomous highway driving feature in 2028 that will allow drivers to have their hands and eyes off the road and without the need to be ready to take over. The software race includes hardware. European and U.S. automakers have spent years chasing the prospect of software-defined vehicles, but have often fallen short. Volkswagen Group had so much trouble with its own in-house effort, including its troubled Cariad unit, that it struck a $5.8 billion joint venture deal with relative newcomer Rivian.

今年，通用汽车提供了更多细节和时间表。这项技术改革预计将于 2028 年随凯迪拉克 Escalade IQ 首次亮相。该公司还将在 2028 年推出一项高速公路自动驾驶功能，允许驾驶员在无需随时准备接管的情况下，实现双手脱离方向盘、双眼离开路面。软件竞赛也涵盖了硬件。欧美汽车制造商多年来一直追求“软件定义汽车”的前景，但往往收效甚微。大众集团在内部研发方面遇到了巨大困难，包括其陷入困境的 Cariad 部门，以至于不得不与相对较新的 Rivian 达成了一项 58 亿美元的合资协议。

GM has been on its own capitally intensive software adventure. The company has bet heavily on software in recent years. It ditched the popular smartphone projection technology Apple CarPlay for nearly all of its EVs, and instead turned to a Google-based infotainment system, which is actually quite capable even if it doesn’t yet live up to the standards set by Tesla and Rivian. It’s poured billions into software and autonomous vehicle technology development too. GM’s capitalized software amortization expense in 2025 was $866 million, which gives a sense of how much it is spending even as it spreads out the cost over years.

通用汽车也一直在进行其资本密集型的软件探索。近年来，该公司在软件领域投入巨大。它在几乎所有电动汽车中放弃了流行的智能手机投屏技术 Apple CarPlay，转而采用基于谷歌的信息娱乐系统。尽管该系统尚未达到特斯拉和 Rivian 的标准，但其实际表现相当出色。通用汽车还在软件和自动驾驶技术开发上投入了数十亿美元。2025 年，通用汽车的资本化软件摊销费用为 8.66 亿美元，这反映出即便将成本分摊到多年，其投入依然巨大。

Like many legacy automakers, GM has come to the conclusion that the old approach to electronics, which scattered computer chips throughout the vehicle, was holding it back. For fourth-generation EVs, GM will centralize much of those functions on a Central Compute Unit (CCU) that’s 40-times more powerful than its existing systems. Eliminating smaller chips on the edges will help trim wiring by 30%, saving weight and simplifying assembly. GM notes that this new CCU will work on fossil fuel models, too, though it will debut in the Cadillac Escalade IQ, a vehicle that is shaping up to be the automaker’s technological flagship.

像许多传统汽车制造商一样，通用汽车得出结论：那种将计算机芯片分散在车辆各处的旧电子架构正在阻碍其发展。对于第四代电动汽车，通用汽车将把大部分功能集中在一个中央计算单元（CCU）上，其性能比现有系统强大 40 倍。消除边缘的小型芯片将有助于减少 30% 的布线，从而减轻重量并简化组装。通用汽车指出，这种新的 CCU 也将适用于化石燃料车型，尽管它将首先在凯迪拉克 Escalade IQ 上亮相，该车型正逐渐成为该汽车制造商的技术旗舰。

The CCU will handle the usual propulsion and chassis controls, but also perception and active safety. That should help pave the way for more advanced autonomy features, which GM also announced on Wednesday. Such a move was perhaps inevitable. Pressure on automakers has been building on all sides. Domestically, Tesla has been pushing the boundaries of autonomy, in part because it developed a true software-defined vehicle. Automakers like GM are also feeling pressure from Chinese manufacturers, which have shortened development cycles to two to three years, down from the usual four to five years. GM intends to match that cadence.

CCU 不仅将处理常规的动力和底盘控制，还将负责感知和主动安全。这应有助于为更先进的自动驾驶功能铺平道路，通用汽车也在周三宣布了这一点。此举或许是必然的。汽车制造商正面临来自各方的压力。在国内，特斯拉一直在推动自动驾驶的边界，部分原因在于它开发了真正的软件定义汽车。像通用汽车这样的制造商也感受到了来自中国制造商的压力，后者已将开发周期从通常的四到五年缩短至两到三年。通用汽车打算跟上这一节奏。

Alongside announcements about battery chemistries and autonomy, GM outlined how enterprise AI tools have sped up vehicle development across every sector as it pushes to slash its traditional four-to-five-year development timeline in half. GM gave a brief glimpse of its revamped development strategy. One example: engineers went through 300,000 virtual designs of a new lithium-manganese-rich battery pack before they assembled the first physical prototype. That helped the concept move to an approved program in one month. Such speed will help GM save costs as it races to preserve profitability amid a historic upheaval in the automotive industry.

在宣布电池化学成分和自动驾驶的同时，通用汽车还概述了企业人工智能工具如何加速了各个领域的车辆开发，旨在将其传统的四到五年开发周期缩短一半。通用汽车简要展示了其改进后的开发策略。一个例子是：工程师在组装第一个物理原型之前，对一种新型富锂锰电池组进行了 30 万次虚拟设计。这使得该概念在短短一个月内就转变为获批项目。这种速度将帮助通用汽车在汽车行业历史性动荡中保持盈利的同时节省成本。

But will it be enough to catapult it back into the leadership position it once enjoyed? The rise of Chinese automakers won’t make it easy. Their factories are capable of producing far more than Chinese consumers can absorb, and coupled with government subsidies and cheaper labor, these vehicles have flooded markets from Europe to South America with inexpensive vehicles. Chinese automaker’s cost advantage is so steep that, without trade barriers, “they will pretty much demolish most other companies in the world,” Elon Musk said in a 2024 earnings call. In response, he has been trying to pivot Tesla from an automotive company to an AI and robotics company. GM’s move into autonomy signals a similar shift. But to deliver true autonomy, GM realized it didn’t have the platform it needed. With a new architecture, it should have a fighting chance.

但这足以让它重回曾经享有的领导地位吗？中国汽车制造商的崛起让这一切变得并不容易。他们的工厂产能远超中国消费者的需求，再加上政府补贴和更廉价的劳动力，这些车辆已以低廉的价格涌入从欧洲到南美洲的市场。埃隆·马斯克在 2024 年的财报电话会议上表示，中国汽车制造商的成本优势如此巨大，以至于如果没有贸易壁垒，“他们几乎会摧毁世界上大多数其他公司”。作为回应，他一直试图将特斯拉从一家汽车公司转型为人工智能和机器人公司。通用汽车进军自动驾驶领域也标志着类似的转变。但通用汽车意识到，要实现真正的自动驾驶，它此前缺乏所需的平台。有了新的架构，它应该有了一线生机。