Manufacturers face critical cybersecurity challenges as they attempt to integrate artificial intelligence (AI) into their operations. A recent study by TXOne reveals that over half of operational technology (OT) environments still rely on legacy systems, which were never designed to support AI-driven automation. This reliance leaves manufacturers increasingly vulnerable to cyberattacks, particularly as supply chain attacks have surged by 20% year-on-year, according to IDS-INDATA analysis. The National Cyber Security Centre (NCSC) has cautioned that AI will likely enhance the effectiveness of cyber intrusion operations, making it both an operational advantage and a potential adversary for manufacturers.
As manufacturers look to modernize their networks, they must prioritize four key areas to safely harness AI’s productivity benefits while mitigating risks. First, ensuring complete asset visibility and data integrity is essential. AI systems depend on accurate and comprehensive data from all connected devices and operational conditions. Inadequate visibility can lead to production errors and system failures, with nearly 70,000 OT devices exposed to the public internet running outdated firmware with known vulnerabilities. Manufacturers must invest in comprehensive OT and IT asset discovery, real-time system monitoring, and strict data quality controls.
Secondly, embedding security across IT and OT is paramount to support autonomous systems. Legacy networks were not designed with security in mind, and without early integration of robust security measures, manufacturers risk exposure to AI-powered malware and precision attacks. Key steps include network segmentation, secure remote access, continuous threat detection, and ongoing vulnerability management. This is particularly important as many legacy factories were not built for the demands of modern connectivity and cybersecurity.
Thirdly, manufacturers must establish strong governance, compliance, and AI auditability frameworks. As AI systems introduce new responsibilities around data handling and decision-making transparency, a lack of robust governance can lead to regulatory breaches and legal liabilities. Structured AI governance, model traceability, and compliance frameworks aligned with industry standards are necessary to navigate the complexities introduced by AI. UK regulators emphasize the importance of accountability throughout the AI life cycle, making governance and auditability essential.
Finally, building resilience against emerging AI-enabled cyber threats is critical. Cybercriminals are increasingly leveraging AI to enhance their attacks, making it imperative for manufacturers to adopt a layered approach to cybersecurity. Effective resilience measures include incident response planning, protected backups, micro-segmentation, and anomaly detection. The NCSC underscores that AI is making reconnaissance operations more efficient, intensifying the urgency for proactive measures to safeguard industrial systems.
Modernizing networks is not simply a barrier to AI adoption; it represents an opportunity to unlock significant benefits. With the right foundation in place—such as visibility, embedded security, and comprehensive governance—manufacturers can realize the productivity gains AI offers while minimizing associated cyber and compliance risks. Those already reassessing their operational technology security through measures like segmentation and real-time monitoring are positioned to deploy AI with greater confidence, ensuring they can capture the advantages of this transformative technology without jeopardizing their operations.
See also
State Laws Surge as Trump’s AI Order Exempts Data Center Regulations
Dr. Adaeze Oreh Transforms Rivers State Healthcare with $5.1M Funding, 134% Contraceptive Visit Increase
MHRA Launches Call for Evidence on AI Regulation Amid Healthcare Transformation
Trump Announces AI Executive Order to Ensure US Global Dominance and Minimize State Regulations
AI Use in Employment Decision-Making Surges to 80% Amid Legal Risks and Compliance Concerns
