Honeywell Survey Finds OT Security AI Adoption High, Autonomous Use Rare

Summary

Honeywell’s 2026 OT Cybersecurity Benchmark Report, based on a survey of 603 leaders, found a gap between perceived program maturity and security visibility: 88% called their OT programs mature or design-led, but only 21% reported a complete OT asset inventory. AI tools are widely used in OT security, while autonomous or agentic AI remains uncommon.

Key points

• Only 21% of respondents reported a complete inventory of OT assets, despite 88% rating their programs mature or design-led.
• OT was fully integrated into a centralized security operations center for 33% of respondents; only 20% continuously monitored more than three-quarters of connected IoT devices.
• Organizations reporting a significant OT incident averaged 16.2 hours of downtime; some incident-affected respondents estimated costs above $100,000 per hour.
• In the past 12 months, significant OT incidents were reported by 91% of energy and utilities respondents, 87% of maritime respondents, and 54% of oil and gas respondents.
• AI use was reported for threat detection by 72%, continuous monitoring by 68%, and asset inventory by 59% of respondents.
• Only 23% currently use autonomous or agentic AI for threat detection; Honeywell emphasized human oversight, clear decision rights, and testing for operational consequences.

Tags

Industries