影响因素-如何利用人工智能实现可持续和可扩展的业务运营

How to leverage AI forsustainable and scalable In Summary •By driving digital transformation through artificial intelligence (AI), organizations can strengthen resilience, •Organizations across the world are facing an enormous challenge: balancing sustainability withcompetitiveness, in an increasingly uncertain context. While AI can deliver enormous business value, it can •However, when implemented strategically with proper guardrails and a robust human-AI framework,agentic AI enables a new model for creating value while embedding sustainability into core business The degradation of our planet – through waterwaste, biodiversity loss, climate change, and otherdisruptions – is one of the greatest threats tohumanity today. Shifting environmental conditions agents – powered by data, generative AI, andadvanced algorithms – research, analyze, and act Though eager to implement AI, organizations areconcerned about its potentially negative impacton the climate. Our most powerful AI – agentic AI –can in fact be put to work to support sustainabilitythroughout organizations. It can boost everythingfrom planning to reporting, with applications Agentic AI offers a breakthrough in sustainability,reshaping digital transformation by enablingorganizations to pursue sustainable practicesmore effectively, efficiently, and at scale. These 1.From automation toautonomy: redefining AI has unfolded in waves, each expanding whatmachines can do. Early AI systems were designed forprecision and repeatability within narrowly definedtasks. Expert systems in medicine, for example,applied symbolic logic to diagnose diseases basedon predefined rules, while early spam filters relied direction, generative AI was the next step towardsautonomous action in AI. By training large languagemodels (LLMs) on massive datasets, developersenabled machines to create original outputs –text, images, or code – through natural languageinterfaces that human managers could easilyunderstand. Tools like Microsoft Copilot brought this The emergence of generative AI broke throughthese boundaries. Flexible, able to draw on Agentic AI emerges as apowerful enabler For maximum impact, businesses need a toolthat can go beyond traditional automation andgenerative AI. This tool must reason, collaborate The building blocks of agentic AI are AI agents:intelligent software systems that perceive theirenvironment, reason, and act autonomously toachieve objectives. Equipped with persistent Agentic AI goes further, orchestrating a network ofagents with memory, planning, and collaborationto enable agents to learn across interactions, self-correct, and work in multi-agent systems. Unlike By moving beyond task execution to goal-oriented collaboration, this ‘agentification’ isredefining enterprise intelligence and architecture, A clear, expanding return on investment AI is already delivering measurable businessvalue, with clear gains in efficiency, agility, andresilience across key areas such as supply chain,manufacturing, and environmental, social, andgovernance (ESG) compliance and operations. Andagentic AI has the potential to go even further.Many organizations have conducted pilot projects, The data is inspiring further action – Capgeminiis finding a clear shift from experimentation to strategic investmentivas confidence in AI’scommercial viability grows. Around 40% oforganizations surveyed anticipate positive ROI These expectations are already translating intotangible benefits across business functions,particularly in procurement, manufacturing, and ESG In manufacturing, for example, digital twins andvirtualized research and development reduce theneed for physical builds, cutting waste, costs andemissions while continuously optimizing production.In logistics, agentic AI can self-correct and optimizelast-mile delivery, lowering costs and reducing The greatest long-term value, however, comeswhen agentic AI is embedded across end-to-end workflows. By integrating AI into corebusiness architectures, organizations can unlockcompounding benefits: optimizing resourceallocation, automating compliance and sustainability Sustainabilitychallenges persist in the2. Organizations face increasing pressure to meetsustainability commitments while contending withthe environmental costs of digitization, supply chain By optimizing energy use, improving supply chaintransparency, and enabling data-driven strategy, However, while both AI implementation and thewider sweep of digitization are often seen asdrivers of efficiency, they can bring steep When implemented strategically and withproper guardrails, agentic AI can drive digital Digitization drives energy use While not all digital services require large-scale dataprocessing, many do – especially AI. In 2024, theInternational Energy Agency (IEA) reported that U.S. data center investments had doubled in two years,viiwhile in Ireland they already accounted for 20% of The digital transformation is deeply intertwinedwith