2026年工业AI状况报告：制造业篇

Introduction Over 350 manufacturing decision makersparticipated in the 2026 State of IndustrialAIReport. Following the success of the inaugural 2024 State of Industrial NetworkingReport, this edition looks at how manufacturing firms worldwide are adoptingArtificial Intelligence; the challenges they face; and the opportunities forAI-powered transformation. We spoke to decision-makers at firms in19 countries, operating with annual revenues over $100 million. This sector report provides a manufacturing-specific view of the 2026 State of Industrial AIresearch, highlighting where the sector is seeingprogress, where constraints remain, and whatfoundations are required to move from targeteddeployments to secure, repeatable scale. Cisco, in association with Sapio Research, undertook this study to establish levelsof AI adoption in industrial networking; the operational outcomes organizations areachieving from AI investments; barriers to scale; and how to align for success. Executive summary Industrial AI demands networkmodernization Cybersecurity is both the #1 barrierand the #1 asset IT/OT collaboration is critical to AIat scale Expanding AI adoption elevates cybersecurity risk. Effective collaboration between IT and OTteams directly impacts AI outcomes. AI implementation is placing unprecedenteddemands on manufacturers’ underlying infrastructure. ●40% cite cybersecurity concern asthe top barrier to AI adoption●46% say security is the biggest networkingchallenge for AI-enabled operations●81% expect AI to improve theircybersecurity posture ●56% say unreliable wireless connectivityfrequently disrupts manufacturing operations●96% say wireless connectivityis critical to AI success●Reliable connectivity (49%), edge compute(44%), and bandwidth (39%) are topnetwork requirements for AI at scale ●43% of manufacturing organizationsshow little to no IT/OT collaboration●34% cite lack of collaboration betweenIT and OT teams as a major challengelimiting AI-enabled operations●28% say OT domain expertiseis critical to scaling AI While security gaps are limiting AI scale today,manufacturers view AI as a tool to strengthendetection, monitoring, and resilience. While disparate teams slow AI deploymentand increase operational risk, IT/OT alignmentaccelerates scalability, stability, and security. Network readiness now determines AIsuccess, with infrastructure limitations directlyconstraining the ability to scale deployments. Table of Contents Section 4: Cybersecurity & industrialAIinterconnectedness21Cybersecurity is now the #1 obstacle toAIadoption22Cybersecurity as a foundational requirement23Cybersecurity threats of greatest concern24AIstrengthens cyber defense25Section 5: IT/OT convergence – the operating model for scalingAI26IT/OT collaboration in manufacturing remains uneven27Independent teams limitAIconfidence & outcomes28Skills required to scale industrialAI29Section 6: Future outlook – scaling AI across industrial operations30Confidence in scaling industrial AI31How transformational willAIbe for manufacturing?32Section 7: Key takeaways for manufacturing leaders33Priorities for manufacturing leaders34Section 8: Industrial AI partner considerations for manufacturing35Section 9: Demographics & firmographics37 Introduction2Executive summary3Introduction letter5Section 1: The state of industrial AI adoption in manufacturing6AI adoption has reached active deployment7Operational improvements driveAIadoption8AIadoption evolves from efficiency to resilience9Cybersecurity concerns limitAIadoption10Section 2:AIoutcomes, ROI & investment urgency11Manufacturers expectAIto deliver operational gains12AI investments bring high expectations13Investment priorities for enablingAI14AIinvestment priorities shift with maturity15Section 3:AIruns on the network: infrastructure as the foundation for scale 16AIadoption rewrites industrial infrastructure requirements17Network readiness is the primary constraint toAIscale18PoweringAIat the edge: PoE as a scaling constraint in manufacturing19ScaledAIadoption requires a network evolution20 Introduction letter I’m excited to introduce Cisco’s 2026 State of Industrial AIReport – an evolution of our State of Industrial NetworkingReport. As industrial operations continue to transform, artificialintelligence is rapidly emerging as a gamechanger acrossmanufacturing, utilities, transportation, and beyond. 2025 wasthe year for many to experiment with AI in industrial settings;2026 promises to be the year when many organizationsmove from pilots to real, production-ready AI projects. We’re seeing companies bring AI to life in impactful ways:from deploying machine vision to ensure product quality inmanufacturing, to rolling out AI-powered automated guidedvehicles (AGVs) and autonomous mobile robots (AMRs) that arereshaping material handling and logistics, to leveraging agenticoperations that drive more autonomous, adaptive, and efficientworkflows across industrial environments. Additionally, AI