Industrial organizations are rapidly increasing their use of artificial intelligence (AI) across sectors such as manufacturing, utilities, and transportation. However, this trend is accompanied by significant security concerns. According to Cisco’s 2026 State of Industrial AI Report, which surveyed over 1,000 decision-makers from 19 countries, cybersecurity has emerged as the primary barrier to AI adoption, surpassing issues like skills shortages, integration difficulties, and budget constraints.
This shift in priorities is noteworthy. In 2024, cybersecurity was ranked third among external growth challenges. By 2026, 40% of respondents identified it as a leading obstacle to AI adoption, with 48% citing it as their biggest networking challenge overall. This trend highlights the reality that connecting more assets and systems to support AI increases the attack surface significantly, posing challenges that traditional security measures were not designed to address.
Widespread Deployment but Limited Transformation
Most organizations are not merely experimenting with AI; they are deploying it on a large scale. The report reveals that 61% of organizations are actively implementing AI, with a significant number executing projects across multiple sites. Only 14% remain in the exploratory or pilot phases.
The motivations for AI adoption are primarily operational, with productivity enhancement and cost reduction leading the charge. An impressive 87% of respondents believe they will see results within two years. This short timeframe encourages organizations to pursue use cases that can quickly demonstrate value, such as process automation and quality inspection.
Infrastructure Readiness as a Limiting Factor
As AI workloads increase, they demand more from infrastructures that many industrial networks were not designed to support. Most decision-makers emphasize the importance of reliable wireless networks for enabling AI, with half expecting substantial increases in connectivity and reliability needs as deployments grow. Alarmingly, 48% of respondents report that security and segmentation challenges are their most significant networking concerns, indicating that infrastructure and security issues are often intertwined.
Currently, AI accounts for 13% of networking budgets, and 83% of organizations plan to boost this allocation. Investments in edge computing, AI vision systems, and industrial connectivity are among the highest priorities as companies transition from human-in-the-loop workflows to machine-to-machine decision-making.
Persistent IT/OT Gaps Affecting Outcomes
Collaboration between IT and operational technology (OT) teams remains inconsistent. Forty-three percent of organizations report limited or no cooperation between these teams, a statistic that has not improved since 2024. This lack of collaboration has tangible consequences; 90% of organizations with siloed teams experience wireless instability, compared to just 61% of those with more integrated structures. Moreover, confidence in scaling AI correlates closely with organizational alignment.
Samuel Pasquier, Product Management Lead for Cisco Industrial IoT Networking, attributes these challenges to discipline gaps rather than a lack of motivation. He stated, “The biggest barrier to IT/OT collaboration is the reality that IT and OT come from very different disciplines, with different technologies, knowledge, priorities, and definitions of risk. Expecting individuals to span both worlds is unrealistic; what matters is enabling collaboration, not convergence of roles.”
As AI systems transition from pilot programs to full production, effective coordination becomes increasingly critical. In environments where silos exist, organizations find it challenging to deploy AI confidently, regardless of technological advancements. Progress varies by sector, with the Hi-Tech Electronics and Semiconductor industry leading; 64% of organizations in this sector report high confidence in scaling AI, followed by energy and transportation sectors. Some regions in Europe are progressing faster, viewing AI as a shared operational capability.
Only 20% of organizations report fully collaborative IT/OT efforts concerning cybersecurity. This gap is significant in terms of risk perception. The report indicates that organizations with higher collaboration are more likely to identify cybersecurity as a primary obstacle, at a rate 12 percentage points higher than those that operate independently. This difference is attributed to enhanced visibility; collaborative efforts often reveal risks that isolated teams might overlook.
AI as Part of the Security Solution
Although cybersecurity presents a significant hurdle, organizations are heavily investing in AI to bolster their defenses. Eighty-five percent of respondents anticipate that AI will enhance their cybersecurity posture, making industrial cybersecurity the second most crucial area for AI investment overall. The expectation is that AI will improve detection, monitoring, and response capabilities at a scale and speed that manual methods cannot achieve.
For manufacturers still reliant on legacy infrastructure, Pasquier emphasizes that modernization does not necessarily require replacing existing systems. The first priority is achieving visibility. “You cannot protect or feed data to an AI if you don’t know what’s on your wire,” he explained. This visibility should extend beyond north-south communication to encompass east-west traffic between devices, utilizing network telemetry as a practical solution. The next step is network segmentation to safeguard AI workloads, ensuring that security incidents on the office network do not impact the plant floor. Finally, organizations should move toward unified IT/OT governance, treating OT cybersecurity as a shared baseline.
The report projects a mixed outlook for the next three to five years, reflecting both confidence and caution. While 93% of organizations feel assured about their ability to scale AI, only one-third expect to see comprehensive operational transformation during this timeframe. Most are using AI to enhance existing processes rather than fundamentally redesigning their operations.
The disparity between confidence and transformation highlights the intersection of infrastructure, security, and organizational structure. Organizations that are most advanced in AI deployment share common traits: modernized networks, mature cybersecurity practices, and collaborative IT/OT governance. These conditions are not yet widespread, suggesting that industrial-scale AI adoption will remain an exception until they are more commonly established.
