Security Testing (7dc672f6-16af-5276-8266-6f5d0b43106d)

AI systems can be targeted in unique ways, such as adversarial inputs, poisoning attacks, or reverse-engineering of model outputs. These threats could compromise the system's confidentiality, integrity, and availability, leading to reputational damage or harm to users. Testing for these issues may require specialized expertise, tools, and time, which could affect project timelines.

Threat-modeling question: Could we be deploying the AI system without testing for adversarial robustness and systemic vulnerabilities?

Cluster A	Galaxy A	Cluster B	Galaxy B	Level
Security Testing (7dc672f6-16af-5276-8266-6f5d0b43106d)	PLOT4ai	Validate ML Model (01c2ec0a-e257-4a75-9e59-f71aa6362b6e)	MITRE ATLAS Course of Action	1
Model (452b8fdf-8679-4013-bb38-4d16f65430bc)	MITRE ATLAS Attack Pattern	Validate ML Model (01c2ec0a-e257-4a75-9e59-f71aa6362b6e)	MITRE ATLAS Course of Action	2
Validate ML Model (01c2ec0a-e257-4a75-9e59-f71aa6362b6e)	MITRE ATLAS Course of Action	Poison ML Model (e0eb2b64-aebd-4412-80f3-b71d7805a65f)	MITRE ATLAS Attack Pattern	2
Validate ML Model (01c2ec0a-e257-4a75-9e59-f71aa6362b6e)	MITRE ATLAS Course of Action	Inject Payload (a50f02df-1130-4945-94bb-7857952da585)	MITRE ATLAS Attack Pattern	2
Model (452b8fdf-8679-4013-bb38-4d16f65430bc)	MITRE ATLAS Attack Pattern	ML Supply Chain Compromise (d2cf31e0-a550-4fe0-8fdb-8941b3ac00d9)	MITRE ATLAS Attack Pattern	3
Poison ML Model (e0eb2b64-aebd-4412-80f3-b71d7805a65f)	MITRE ATLAS Attack Pattern	Backdoor ML Model (c704a49c-abf0-4258-9919-a862b1865469)	MITRE ATLAS Attack Pattern	3
Inject Payload (a50f02df-1130-4945-94bb-7857952da585)	MITRE ATLAS Attack Pattern	Backdoor ML Model (c704a49c-abf0-4258-9919-a862b1865469)	MITRE ATLAS Attack Pattern	3