
Manufacturing organisations are increasing their adoption of artificial intelligence (AI), but the growing use of the technology is also creating new cybersecurity and data protection risks, according to a report by Netskope Threat Labs.
The Netskope Threat Labs Report: Manufacturing 2026 found that the proportion of manufacturing employees actively using AI applications rose from 40 per cent to 67 per cent over the past year.
The report, based on anonymised data collected through the Netskope One platform between June 2025 and August 2026, also found that the use of organisation-managed AI applications doubled from 35 per cent to 70 per cent. Meanwhile, the use of personal AI accounts declined from 75 per cent to 55 per cent.
Microsoft Copilot was the most widely adopted AI application among manufacturing organisations, with 90 per cent reporting its use, followed by Claude Platform at 89 per cent and ChatGPT at 76 per cent.
However, the findings also identified concerns about the handling of sensitive information through AI applications. Regulated data accounted for 57 per cent of AI-related data policy violations in manufacturing, followed by passwords and access keys at 19 per cent, intellectual property at 16 per cent and source code at nine per cent.
AI adoption was also extending beyond dedicated applications, with 96 per cent of employees using applications with embedded AI features and 90 per cent interacting with systems that use customer or user data for model training.
The report highlighted the rapid growth of connections between AI agents and remote Model Context Protocol (MCP) servers. The number of non-human entities connecting to these servers increased by 490 per cent, while related events rose by 600 per cent over the research period.
Netskope Threat Labs also identified an increase in malicious activity targeting AI users. The number of manufacturing employees encountering malicious AI lures reached nearly 175 per 100,000 users in June 2026, while interactions with malicious links from AI applications reached 200 per 100,000 users in March 2026.
The report said these developments highlight the need for manufacturers to strengthen visibility over AI applications, manage access to sensitive information and improve security controls across business systems.
Its recommendations include monitoring web and cloud downloads, restricting access to applications that pose security risks, and applying data loss prevention policies to protect regulated information, intellectual property, passwords and source code.
The report draws on aggregated usage data from a subset of Netskope customers in the manufacturing sector, covering automotive, chemicals, consumer goods, food and beverage, aerospace and defence, and heavy industrial engineering.




















