Join our Newsletter — 33% off our NHI Course
Home› Guides› ATT&CK for ICS Inhibit Response Function Tactic Guide:…
Guide OT & IoT Security

ATT&CK for ICS Inhibit Response Function Tactic Guide: Protect Safety, Alarms and Operator Response

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 27 October 2026 14 min read
On this page

Inhibit Response Function is the ATT&CK for ICS tactic in which an adversary stops safety systems, alarms and operators from reacting to a hazard, so that a physical attack can run to completion. MITRE describes the tactic (TA0107) as the adversary "trying to prevent your safety, protection, quality assurance, and operator intervention functions from responding to a failure, hazard, or unsafe state". It is the tactic that turns a process upset into a possible safety incident, because the layers designed to catch the upset are the ones being disabled. This guide walks through every technique in the tactic as published in ATT&CK v20, the mitigations MITRE maps to them, how to detect them in OT telemetry, and how to prioritise defences. It is written for OT security engineers, control system engineers, process safety teams and SOC analysts who cover industrial sites. It builds on the MITRE ATT&CK for ICS Guide and does not repeat the matrix overview.

Key takeaways

  • The tactic targets dormant safeguards. Protection functions sit idle until something goes wrong, so reporting can look normal while the safety layer has already been altered.
  • Fourteen techniques in v20. They range from Alarm Suppression (T0878) and Modify Alarm Settings (T0838) to Modify Firmware (T1693), Block Communications (T1695) and Change Credential (T0892).
  • Engineering access is the common path. Most techniques need a write to a controller, a firmware transfer or a protocol command, so restricting and monitoring engineering workstations and their credentials removes most of the attack surface.
  • Independence is the core defence. Separate safety networks, out-of-band alarm paths (M0810), static network configuration (M0814) and hardwired trips keep a compromised control network from silencing protection.
  • Detect the absence of signals. Alarms that go quiet, alarm states that change with no operator action and devices entering firmware update mode are the observable symptoms, so baseline them.

Where Inhibit Response Function sits in an ICS intrusion

MITRE notes that, unlike Evasion (TA0103), Inhibit Response Function techniques "may be more intrusive, such as actively preventing responses to a known dangerous scenario", and that adversaries use them "to follow through with or provide cover for" Impact (TA0105). In practice the tactic usually appears late in an intrusion, after the adversary has reached an engineering workstation or a controller network and has learned the process. It pairs closely with Impair Process Control (TA0106): one tactic pushes the process into an unsafe state, this one stops the system from noticing or responding. See the ATT&CK for ICS Impair Process Control Tactic Guide.

  • Safety functions: safety instrumented systems (SIS), emergency shutdown systems and burner management, which trip the process to a safe state. See the Safety Instrumented Systems Security Guide.
  • Protection functions: protective relays in power systems and interlocks in PLC logic.
  • Alarm and reporting functions: HMI alarms, historian events and SCADA alarm servers that tell operators something is wrong.
  • Operator intervention: the ability of staff to reach the HMI, log in, send a command or take manual control.

The Triton incident (ATT&CK software S1009, made public in December 2017) is the reference case. MITRE records that Triton was built to interact with Schneider Electric Triconex SIS controllers, used the TriStation protocol on UDP port 1502, wrote to controller firmware memory (System Firmware, T1693.001), added code to the program table (Modify Controller Tasking, T0821) and could reprogram SIS logic so that unsafe conditions persist (Loss of Safety, T0880).

The techniques in ATT&CK v20

The table lists every technique MITRE shows under TA0107 in ATT&CK v20, with examples of software and campaigns that ATT&CK links to each one. Several IDs changed in v20 (for example Block Communications T1695 and Modify Firmware T1693 replace older single techniques), so update detection mappings that still use the old IDs.

IDTechniqueWhat it does to the response layerExamples recorded by ATT&CK
T0800Activate Firmware Update ModePlaces a device in update mode so it stops performing its normal response functionIndustroyer
T0878Alarm SuppressionPrevents protection alarms from reaching operatorsMaroochy Water Breach
T1695 (.001 Serial COM, .002 Ethernet, .003 Wi-Fi)Block CommunicationsCuts the medium that carries commands and reportsIndustroyer, LockerGoga, CyberAv3ngers
T1691 (.001 Command, .002 Reporting)Block Operational Technology MessageDrops specific command or telemetry messagesIndustroyer, 2015 Ukraine Electric Power Attack
T0892Change CredentialLocks operators and responders out of devicesCyberAv3ngers, 2025 Poland Wiper Attacks
T0809Data DestructionDeletes files, projects or backups needed for recoveryKillDisk, Industroyer, INCONTROLLER
T0814Denial of ServiceOverwhelms or crashes a device so it cannot respondIndustroyer, PLC-Blaster, Fuxnet
T0816Device Restart/ShutdownForces a reboot or shutdown during an upsetIndustroyer, 2015 Ukraine Electric Power Attack
T0835Manipulate I/O ImageOverrides the PLC input/output image so logic acts on false valuesStuxnet, PLC-Blaster
T0838Modify Alarm SettingsChanges alarm thresholds or disables alarmsCyberAv3ngers, Maroochy Water Breach
T1693 (.001 System, .002 Module)Modify FirmwareReplaces or patches device firmwareTriton, FrostyGoop Incident
T0851RootkitHides malicious code and changes from engineering toolsStuxnet
T0881Service StopStops services such as historians, HMIs or protocol gatewaysEKANS, Industroyer2, REvil

Note that T1693 Modify Firmware is shared with Persistence and Impair Process Control, and T0851 Rootkit is shared with Evasion, so one control often covers several tactics.

Mitigations MITRE maps to the tactic

ATT&CK for ICS links each technique to mitigations. Grouping them shows that a small set of controls covers most of the tactic.

MitigationTechniques it is mapped toWhat it looks like on site
M0930 Network SegmentationT0800, T0816, T0838, T1693, T1695, T0881, T0878Separate SIS and safety networks from the basic process control system; conduits only through defined boundaries
M0807 Network AllowlistsT0800, T0816, T0838, T1691, T1693, T1695, T0878Controllers and switches accept engineering traffic only from named engineering workstations
M0810 Out-of-Band Communications ChannelT0878, T1691, T1695Independent alarm path, such as hardwired annunciators or a separate radio or telephone channel
M0814 Static Network ConfigurationT0878, T1691Static ARP and addressing on critical segments so traffic cannot be redirected
M0804 Human User Authentication and M0800 Authorization EnforcementT0800, T0816, T0838, T1693Logins and role checks on HMIs, engineering software and controllers that support them
M0813 Software Process and Device Authentication and M0802 Communication AuthenticityT0800, T0816, T0838, T1693Authenticated protocols, for example OPC UA with signing or DNP3 Secure Authentication
M0945 Code Signing and M0946 Boot IntegrityT1693, T0851Devices that accept only vendor-signed firmware and verify it at boot
M0815 Watchdog TimersT0814Watchdogs that restart a hung device into a known state
M0953 Data Backup and M0811 Redundancy of ServiceT0809, T0892Offline copies of controller projects and credentials; redundant controllers and servers
M0927 Password Policies and M0918 User Account ManagementT0892, T0838, T0881Unique device passwords, recorded in a vault, with defined owners

MITRE marks Manipulate I/O Image (T0835) with M0816 "Mitigation Limited or Not Effective": prevention depends on stopping the earlier access and logic changes, so detection and engineering integrity checks matter more there.

Engineering and architecture defences

Process safety engineering already assumes failures happen. Security for this tactic extends that assumption to deliberate failure of the safety and alarm layers.

  1. Keep the SIS independent: put safety controllers on their own network, with no routable path from the business network and a tightly filtered conduit to the control system. IEC 61511 and IEC 62443 both expect this separation. See the OT Network Segmentation, Purdue Model and IEC 62443 Zones and Conduits Guide.
  2. Use physical keyswitches and run modes: many safety controllers and PLCs have a key or mode that must be in program position to accept downloads. Leave them in run, log every change and treat a controller left in program mode as an alarm. Triton relied on controllers that accepted program changes.
  3. Provide a second alarm path: critical trips and alarms should reach operators through a path that does not depend on the HMI network, such as a hardwired panel or separate alarm annunciator (M0810).
  4. Restrict firmware update mode: require local presence, a maintenance window and a named engineer before a device can enter firmware update mode, and block the update protocol at the zone boundary outside those windows.
  5. Protect the alarm configuration: keep a signed baseline of alarm limits and shelving rules in alarm management software and compare it daily with the live system. ISA-18.2 style alarm management makes unauthorised changes visible.
  6. Plan for lockout: keep recovery credentials for each controller and HMI in an offline vault, and test that operators can still take manual control if the HMI is lost.

Where identity fits

Nearly every technique in the tactic requires an authenticated or trusted session: an engineering workstation logged into a controller, a service account on a historian, a vendor remote access session or a shared HMI login. Change Credential (T0892) itself is an identity attack: the adversary changes device passwords so responders cannot get back in, as ATT&CK records for CyberAv3ngers and the 2026 U.S. Water Utilities Attacks.

  • Engineering workstation accounts: named accounts only, MFA at the jump host, and no standing access from the IT domain. See the OT and ICS Identity and Access Guide.
  • Service and machine accounts: historian collectors, OPC servers and alarm servers run as service accounts. Inventory them, limit their write rights and rotate their passwords.
  • Device credentials: replace default PLC and HMI passwords, store them in a vault and record who changed them last, so an unexpected credential change is detectable.
  • Vendor and integrator access: time-bound, brokered and recorded sessions. See the OT Remote Access Guide.

Detection and measurement

ATT&CK v20 publishes detection strategies for each technique. For this tactic they share a pattern: a write, mode change or restart from a source that is not an approved engineering workstation, followed by alarms or telemetry going quiet.

TechniqueSignal to monitorData source
T0878 Alarm SuppressionAlarm shelved, disabled or acknowledged with no operator action recorded, near ICS protocol write trafficAlarm server and historian events, OT network monitoring
T0838 Modify Alarm SettingsChanges to alarm limits from outside the approved management system; drift from the alarm baselineAlarm management database, SNMP and protocol logs
T0800 Activate Firmware Update ModeDevice enters update mode from an unapproved source, then its telemetry stopsPassive OT network monitoring, device logs
T1693 Modify FirmwareLarge transfer on the firmware port, reboot, and a firmware hash that does not match the authorised versionNetwork monitoring, asset inventory firmware versions
T0816 Device Restart/ShutdownRestart commands to several devices in a short windowProtocol monitoring, controller diagnostics
T0892 Change CredentialPassword or account change on a device outside a change ticketDevice audit logs where supported, vault records
T0881 Service Stop and T0809 Data DestructionHistorian, HMI or OPC services stopping; mass file deletion on engineering hostsWindows event logs and EDR on OT hosts

See the OT Network Security Monitoring Guide for sensor placement.

Metrics that show the defences work:

  • Percentage of safety controllers on an isolated network with a reviewed conduit (target 100%).
  • Percentage of controllers left in run mode or with keyswitch in run outside maintenance windows.
  • Number of alarm configuration changes without a matching change ticket each month (target zero).
  • Percentage of critical devices whose firmware version is known and matches the approved baseline.
  • Time to detect a test alarm suppression or controller write in a purple team exercise.
  • Percentage of device credentials held in a vault with a named owner.

Implementation roadmap

  1. Map safety and alarm assets: list each SIS, protective relay, alarm server and HMI, the network it sits on and who can write to it. See the OT Asset Visibility Guide.
  2. Close the easy paths: remove routable paths into safety networks, enforce network allowlists for engineering protocols and set controllers to run mode.
  3. Baseline: record firmware versions, controller program hashes and alarm settings. Store the baseline offline.
  4. Monitor in passive mode first: deploy OT network monitoring in listen-only mode, tune alerts for program downloads, mode changes and firmware transfers for several weeks, and agree alert routing with operations before any blocking.
  5. Add authentication where devices support it: enable controller passwords, authenticated protocols and signed firmware on new or upgraded assets, piloting on one non-critical line before wider rollout. Keep the previous configuration so you can roll back if the process is disturbed.
  6. Exercise: run a tabletop and, where safe, a test on a simulator in which alarms are suppressed. Confirm operators recognise silence as a symptom and know the manual response. See the OT Incident Response Guide.

Decision guidance: small sites with a single control network should start with physical controls (keyswitches, hardwired trips, offline backups) and a single passive sensor, which give the biggest gain for the effort. Large operators with many sites should standardise baselines and detection content centrally and use the technique table above to measure coverage per site. Where a device cannot authenticate commands (many legacy PLCs), rely on segmentation, allowlists and monitoring rather than trying to retrofit cryptography.

Framework mapping

DefenceNIST SP 800-53 Rev. 5NIST CSF 2.0ATT&CK for ICS mitigation
Safety network isolationSC-7 Boundary ProtectionPR.IR-01M0930, M0807
Firmware and logic integritySI-7, SI-7(15), CM-14PR.PS-02M0945, M0946
Authenticated device commandsIA-3, SC-8PR.AA-03M0813, M0802
Change control of alarms and programsCM-3, CM-5PR.PS-01M0800, M0947
Monitoring of alarms and protocol writesSI-4, AU-6DE.CM-01, DE.CM-09Detection strategies per technique
Recovery from destruction and lockoutCP-9, SC-24PR.DS-01M0953, M0811

NIST SP 800-82 Rev. 3 provides the OT overlay for SP 800-53 and is the natural companion. See the NIST SP 800-82 OT Security Guide and the IEC 62443 Guide.

Common failure patterns

  • SIS on the control network: integrated safety and control systems save cost but give an adversary on the control network a route to the safety controllers.
  • Controllers left in program mode: a keyswitch left in program or remote position after commissioning means any workstation on the network can download logic.
  • Alarm floods treated as normal: where operators routinely shelve alarms, an adversary's suppression blends in. Alarm rationalisation is a security control as well as a safety one.
  • Monitoring only for activity: SOC rules that alert on new connections miss the more telling signal, a device or alarm that stops reporting.
  • Shared and unrecorded device passwords: when nobody knows the current PLC password, a Change Credential attack goes unnoticed until someone needs to get in.
  • Backups kept on the engineering workstation: Data Destruction (T0809) on that host removes the project files needed to restore the controller.

Practitioner checklist

  • Every safety controller sits on an isolated network with a documented, filtered conduit.
  • Controller keyswitches or modes are set to run, and mode changes generate an alert.
  • Critical alarms have an out-of-band path that survives loss of the HMI network.
  • Firmware update mode is reachable only from named engineering workstations in a maintenance window.
  • Baselines of firmware, controller logic and alarm settings are stored offline and compared regularly.
  • OT monitoring alerts on program downloads, mode changes, firmware transfers and alarm silence.
  • Device credentials are unique, vaulted and owned, with changes tied to change tickets.
  • Detection mappings use the ATT&CK v20 IDs (T1691, T1693, T1695) rather than retired ones.
  • Operators have rehearsed manual response to silent alarms and HMI lockout.

Standards and references

Related NHI Mgmt Group resources: MITRE ATT&CK for ICS Guide · Safety Instrumented Systems Security Guide · OT and ICS Identity and Access Guide · PLC Security Guide

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 27 October 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.