2025技术趋势报告&空间计算、人工智能、IT升级（中文版）

In Deloitte’s 16th annual Tech Trends report,AI is the common thread of nearly everytrend. Moving forward, it will be part of thesubstructure of everything we do.在德勤(Deloitte）的第16份年度科技趋势报告中，人工智能是几乎所有趋势的共同主线。展望未来，它将成为我们一切工作的一部分。 德勤。洞察力 02 . . .Executive summary05 . . .AI everywhere: Like magic, but with algorithms09. . .Spatial computing takes center stage17. . .What’s next for AI?27. . .Hardware is eating the world37. . .IT, amplified: AI elevates the reach (and remit) of the tech function45 . . .The new math: Solving cryptography in an age of quantum53. . .The intelligent core: AI changes everything for core modernization60. . .Breadth is the new depth: The power of intentional intersectionsINTRODUCTIONINTERACTIONINFORMATIONCOMPUTATIONBUSINESS OF TECHNOLOGYCYBER AND TRUSTCORE MODERNIZATIONCONCLUSIONTable of contentsRAEnS02..行动纲要05...尔无处不在：像魔术一样，但有算法介绍互动09...空间计算占据中心舞台17...艾尔的下一步是什么？信息计算27...硬件正在吞噬世界技术商业37...它，被放大了：人工智能提升了技术功能的范围网络和信任45...新数学：在量子时代解决密码学核心现代化53...智能核心：人工智能改变核心现代化的一切结论60...广度是新的深度：有意交集的力量 Executive summary行动纲要 as electricity to daily business and personal lives. Asour team in Deloitte’s Office of the CTO put finishingtouches on Tech Trends 2025, we realized that AI is acommon thread in nearly every trend. We expect thatgoing forward, AI will be so ubiquitous that it will be apart of the unseen substructure of everything we do, and德勤的旗舰技术报告《技术趋势》(TechTrends)操作为日常商业和个人生活的电力。随着我们在德勤（Deloitte)首席技术官办公室的团队对（2025年技术趋势》进行最后的润色，我们意识到，人工智能几乎是每一种趋势的共同主线。我们预计，未来，人工智能将无处不在，它将成为我们所做一切的看不见的底层结构的一部分，我们最终基至不会知道它的存在。 Tech Trends, Deloitte’s flagship technology report,explores the emergence of trends in three elevating forces(interaction, information, and computation) and threegrounding forces (business of technology, cyber andtrust, and core modernization)—all part of our macrotechnology forces framework (figure 1). Tech Trends2025, our 16th trip around the sun, previews a future讨了三种提升力量(交互、信息和计算)和三种基础力量(技术业务、网络和信任以及核心现代化)的趋势，这些都是我们宏观技术力量框架的一部分(图1)。《2025年科技趋势》是我们第16次绕太阳旅行，它预示了人工智能将成为基础的未来 Introduction介绍 efficient choice for all organizational needs. Enterprisesare now considering small language models and open-source options for the ability to train LLMs on smaller,more accurate data sets. Together with multimodalmodels and AI-based simulations, these new types ofAI are building a future where enterprises can find theright type of AI for each task. That includes AI that notonly answers questions but also completes tasks. In thecoming years, a focus on execution may usher in a newera of agentic AI, arming consumers and organizations满足所有组织需求的有效选择。企业现在正在考虑小型语言模型和开源选项，以便能够在更小、更精确的数据集上训练LLM。加上多模态模型和基于人工智能的模拟，这些新型人工智能正在构建一个未来，企业可以为每项任务找到正确的人工智能类型。这包括不仅回答问题而且完成任务的人工智能。在未来几年，对执行的关注可能会迎来一个代理人工智能的新时代，为消费者和组织提供能够改变我们工作和生活方式的副驾驶。 AI everywhere: Like magic, but with algorithms艾尔无处不在：像魔术一样，但有算法 Generative AI continues to be the buzzword of theyear, but Tech Trends 2025—and in fact, the future oftechnology—is about much more than AI. This year’sreport reveals the extent to which AI is being woveninto the fabric of our lives. We’ll eventually take it forgranted and think of it in the same way that we think ofHTTP or electricity: We’ll just expect it to work. AI willperform quietly in the background, optimizing trafficin our cities, personalizing our health care, or creating生殖人工智能仍然是今年的流行语，但2025年技术趋势一一事实上，技术的未来一—不仅仅是人工智能。今年的报告揭示了人工智能在多大程度上融入了我们的生活。我们最终会认为它是理所当然的，并像看待HTTP或电一样看待它：我们只是希望它能工作。人工智能将在后台安静地运行，优化我们城市的交通，个性化我们的医疗保健，或在教育中创建适应性强和可访问的学习路径。 Computation计算 We won’t proactively use it; we’ll simply experience aworld in which it makes everything work smarter, faster,and more intuitively—like magic, but grounded in algo-rithms. The six chapters of Tech Trends 2025 reflect thisemerging reality.我们不会主动使用它；我们将简单地体验一个世界，在这个世界中，它让一切工作得更智能、更快、史直观一一就像魔术一样，但基于算法。（2025年科技趋势》的六个章节反映了这一新兴现实。 Hardware is eating the world硬件正在吞噬世界 After years of software dominance, hardware is reclaim-ing the spotlight. As AI demands specialized computingresources, companies are turning to advanced chips topower AI workloads. In addition, personal comput-ers embedded with AI chips are poised to superchargeknowledge workers by providing access to offline AImodels while “future-proofing” technology infrastruc-ture, reducing cloud computing costs, and enhancingdata privacy. Although AI’s increased energy demandspose sustainability challenges, advancements in energysources and efficiency are making AI hardware moreaccessible. Looking forward, AI’s continued integrationinto devices could revolutionize the Internet of Things在多年的软件统治之后，硬件重新成为焦点。由于人工智能需要专门的计算资源，公司正在转向先进的芯片来支持人工智能工作负载。此外，嵌入人工智能芯片的个人电脑准备通过提供对离线人工智能模型的访问，同时“面向未来”的技术基础设施，降低云计算成本，并增强数据隐私，来增强知识工作者的能力。尽管人工智能增加的能源需求带来了可持续发展的挑战，但能源和效率的进步正在使人工智能硬件变得更容易获得。展望未来，人工暂能继续集成到设备中可能会彻底改变物联网和机器人技术，通过更智能、更自主的设备改变医疗保健等行业。 Interaction互动 Spatial computing takes center stage空间计算占据中心舞台 Spatial computing continues to spark enterprise interestbecause of its ability to break down information silos andcreate more