On Tuesday, Cisco presented the report "The State of Industrial AI – 2026", according to which the majority of industrial organizations have already moved beyond the experimentation phase and are moving towards deploying artificial intelligence in real operational environments, where reliability and safety have very specific consequences. The document outlines the transition from pilot projects to large-scale deployments in critical infrastructure and production.
The annual survey covers over 1,000 operational technology specialists from 19 countries and 21 industrial sectors. According to the results, 61% of organizations are already applying AI in active production processes, with 20% reporting large-scale, mature deployments in areas such as manufacturing, transportation, and utilities. Artificial intelligence is used for process automation, automated quality control, predictive maintenance, logistics optimization, and energy consumption forecasting.
Costs are rising, but readiness gaps remain
Interest in industrial AI continues to grow: 83% of organizations plan to increase spending on AI, and 87% expect tangible results within the next two years. The main motivations for implementation are increased productivity (63%), reduced costs (48%) and improved safety (34%).
At the same time, the report emphasizes that growing readiness gaps threaten to slow progress. Forty percent of the participants cite cybersecurity issues as a major obstacle to AI deployment – a factor that was only in third place in Cisco's 2024 report. Almost half of manufacturing companies report that unstable wireless connectivity negatively affects the performance of AI systems, and many note that edge computing and network capacity continue to be a serious challenge.
"Industrial AI is moving from experiments to real production, where artificial intelligence systems perceive the surrounding world, analyze it and make decisions," commented Vikas Butani, Senior Vice President and General Manager of the Secure Routing and Industrial IoT division at Cisco. "At this stage, success is no longer determined only by the models, but by how ready the networks, security systems and teams are to support AI at the edge, on the go and at scale."
Collaboration between IT and OT is becoming a decisive factor
The report also highlights the strong link between the level of collaboration between IT and OT departments and the organization's confidence that it can scale its AI initiatives. Companies that have well-coordinated work between information technology and operational technology teams more often report a more stable network infrastructure and a higher focus on cybersecurity. Conversely, many organizations still operate with limited interaction or almost complete separation between these two functions, which weakens network efficiency and slows down the deployment of AI.
Cybersecurity, while remaining a major obstacle, is also seen as a key area for the application of artificial intelligence itself. Eighty-one percent of manufacturers state that they rely on AI to improve their cyber defense capabilities after its wider deployment. As noted by Cisco in the report: "although security gaps today limit the scaling of AI, organizations see it as a tool for improving threat detection, monitoring and increasing resilience".