Threat actors may perform data poisoning attacks against the training data sets that are being used for security features driven by artificial intelligence (AI) and/or machine learning (ML). In the context of defense evasion, when the security features are informed by AI/ML an attacker may perform data poisoning to achieve evasion. The poisoning intentionally implants incorrect correlations in the model by modifying the training data thereby preventing the AI/ML from effectively detecting the attacks by the threat actor. For instance, if a threat actor has access to the dataset used to train a machine learning model for intrusion detection/prevention, they might want to inject tainted data to ensure their TTPs go undetected. With the datasets typically used for AI/ML (i.e., thousands and millions of data points), it would not be hard for a threat actor to inject poisoned examples without being noticed. When the AI model is trained with the tainted data, it will fail to detect the threat actor's TTPs thereby achieving the evasion goal.
ID | Name | Description | NIST Rev5 | D3FEND | ISO 27001 | |
CM0049 | Machine Learning Data Integrity | When AI/ML is being used for mission critical operations, the integrity of the training data set is imperative. Data poisoning against the training data set can have detrimental effects on the functionality of the AI/ML. Fixing poisoned models is very difficult so model developers need to focus on countermeasures that could either block attack attempts or detect malicious inputs before the training cycle occurs. Regression testing over time, validity checking on data sets, manual analysis, as well as using statistical analysis to find potential injects can help detect anomalies. | AC-3(11) SC-28 SC-28(1) SC-8 SC-8(2) SI-7 SI-7(1) SI-7(2) SI-7(5) SI-7(6) SI-7(8) | A.8.4 A.5.10 A.5.14 A.8.20 A.8.26 A.5.10 A.5.33 | ||
CM0069 | Process White Listing | Simple process ID whitelisting on the firmware level could impede attackers from instigating unnecessary processes which could impact the spacecraft | CM-7(5) SI-10(5) | A.8.19 | ||
CM0056 | Data Backup | Implement disaster recovery plans that contain procedures for taking regular data backups that can be used to restore critical data. Ensure backups are stored off system and is protected from common methods adversaries may use to gain access and destroy the backups to prevent recovery. | CP-9 | A.5.29 A.5.33 A.8.13 | ||
CM0032 | On-board Intrusion Detection & Prevention | Utilize on-board intrusion detection/prevention system that monitors the mission critical components or systems and audit/logs actions. The IDS/IPS should have the capability to respond to threats and it should address signature-based attacks along with dynamic never-before seen attacks using machine learning/adaptive technologies. The IDS/IPS must integrate with traditional fault management to provide a wholistic approach to faults on-board the spacecraft. Spacecraft should select and execute safe countermeasures against cyber-attacks. These countermeasures are a ready supply of options to triage against the specific types of attack and mission priorities. Minimally, the response should ensure vehicle safety and continued operations. Ideally, the goal is to trap the threat, convince the threat that it is successful, and trace and track the attacker — with or without ground support. This would support successful attribution and evolving countermeasures to mitigate the threat in the future. “Safe countermeasures” are those that are compatible with the system’s fault management system to avoid unintended effects or fratricide on the system. | AU-14 AU-2 AU-3 AU-3(1) AU-4 AU-4(1) AU-5 AU-5(2) AU-5(5) AU-6(1) AU-6(4) AU-8 AU-9 AU-9(2) AU-9(3) CA-7(6) CM-11(3) CP-10 CP-10(4) IR-4 IR-4(11) IR-4(12) IR-4(14) IR-5 IR-5(1) RA-10 RA-3(4) SA-8(21) SA-8(22) SA-8(23) SC-16(2) SC-32(1) SC-5(3) SC-7(9) SI-10(6) SI-16 SI-17 SI-4 SI-4(10) SI-4(11) SI-4(16) SI-4(2) SI-4(25) SI-4(4) SI-4(5) SI-6 SI-7(17) SI-7(8) | A.8.15 A.8.15 A.8.6 A.8.17 A.5.33 A.8.15 A.8.15 A.5.29 A.5.25 A.5.26 A.5.27 A.5.7 A.8.16 A.8.16 A.8.16 | ||
CM0042 | Robust Fault Management | Ensure fault management system cannot be used against the spacecraft. Examples include: safe mode with crypto bypass, orbit correction maneuvers, affecting integrity of telemetry to cause action from ground, or some sort of proximity operation to cause spacecraft to go into safe mode. Understanding the safing procedures and ensuring they do not put the spacecraft in a more vulnerable state is key to building a resilient spacecraft. | CP-4(5) SA-8(24) SC-16(2) SC-24 SI-13 SI-17 | |||
CM0044 | Cyber-safe Mode | Provide the capability to enter the spacecraft into a configuration-controlled and integrity-protected state representing a known, operational cyber-safe state (e.g., cyber-safe mode). Spacecraft should enter a cyber-safe mode when conditions that threaten the platform are detected. Cyber-safe mode is an operating mode of a spacecraft during which all nonessential systems are shut down and the spacecraft is placed in a known good state using validated software and configuration settings. Within cyber-safe mode, authentication and encryption should still be enabled. The spacecraft should be capable of reconstituting firmware and software functions to pre-attack levels to allow for the recovery of functional capabilities. This can be performed by self-healing, or the healing can be aided from the ground. However, the spacecraft needs to have the capability to replan, based on equipment still available after a cyber-attack. The goal is for the spacecraft to resume full mission operations. If not possible, a reduced level of mission capability should be achieved. Cyber-safe mode software/configuration should be stored onboard the spacecraft in memory with hardware-based controls and should not be modifiable. | CP-10 CP-10(4) CP-12 CP-2(5) IR-4 IR-4(12) IR-4(3) SA-8(21) SA-8(23) SA-8(24) SC-16(2) SC-24 SI-11 SI-17 SI-7(17) | A.5.29 A.5.25 A.5.26 A.5.27 |