MITRE ATLAS Attack Landscape

AI attack technique frequency based on 3,786 CVE-to-technique mappings across 97 ATLAS techniques.

Techniques

3,786

CVE Mappings

Exploit Public-Facing Application

#1 (948 CVEs)

#	Technique	Tactic	CVEs
1	AML.T0049 Exploit Public-Facing Application	Initial Access	948
2	AML.T0029 Denial of AI Service	Impact	449
3	AML.T0025 Exfiltration via Cyber Means	Exfiltration	239
4	AML.T0034 Cost Harvesting	Impact	165
5	AML.T0050 Command and Scripting Interpreter	Execution	157
6	AML.T0037 Data from Local System	Collection	155
7	AML.T0055 Unsecured Credentials	Credential Access	141
8	AML.T0053 AI Agent Tool Invocation	Privilege Escalation	124
9	AML.T0072 Reverse Shell	Command and Control	116
10	AML.T0040 AI Model Inference API Access	AI Model Access	115
11	AML.T0058 Publish Poisoned Models	Resource Development	102
12	AML.T0012 Valid Accounts	Privilege Escalation	101
13	AML.T0035 AI Artifact Collection	Collection	93
14	AML.T0083 Credentials from AI Agent Configuration	Credential Access	77
15	AML.T0107 Exploitation for Defense Evasion	Defense Evasion	75
16	AML.T0011 User Execution	Execution	65
17	AML.T0074 Masquerading	Defense Evasion	63
18	AML.T0043 Craft Adversarial Data	AI Attack Staging	59
19	AML.T0006 Active Scanning	Reconnaissance	46
20	AML.T0105 Escape to Host	Privilege Escalation	42
21	AML.T0081 Modify AI Agent Configuration	Persistence	40
22	AML.T0106 Exploitation for Credential Access	Credential Access	38
23	AML.T0078 Drive-by Compromise	Initial Access	33
24	AML.T0075 Cloud Service Discovery	Discovery	22
25	AML.T0020 Poison Training Data	Resource Development	20
26	AML.T0085 Data from AI Services	Collection	19
27	AML.T0086 Exfiltration via AI Agent Tool Invocation	Exfiltration	18
28	AML.T0007 Discover AI Artifacts	Discovery	17
29	AML.T0080 AI Agent Context Poisoning	Persistence	17
30	AML.T0018 Manipulate AI Model	Persistence	16
31	AML.T0051 LLM Prompt Injection	Execution	16
32	AML.T0057 LLM Data Leakage	Exfiltration	13
33	AML.T0031 Erode AI Model Integrity	Impact	11
34	AML.T0052 Phishing	Lateral Movement	11
35	AML.T0101 Data Destruction via AI Agent Tool Invocation	Impact	11
36	AML.T0001 Search Open AI Vulnerability Analysis	Reconnaissance	10
37	AML.T0036 Data from Information Repositories	Collection	9
38	AML.T0076 Corrupt AI Model	Defense Evasion	9
39	AML.T0070 RAG Poisoning	Persistence	8
40	AML.T0084 Discover AI Agent Configuration	Discovery	8
41	AML.T0021 Establish Accounts	Resource Development	7
42	AML.T0044 Full AI Model Access	AI Model Access	7
43	AML.T0059 Erode Dataset Integrity	Impact	7
44	AML.T0073 Impersonation	Defense Evasion	6
45	AML.T0079 Stage Capabilities	Resource Development	6
46	AML.T0093 Prompt Infiltration via Public-Facing Application	Initial Access	6
47	AML.T0097 Virtualization/Sandbox Evasion	Defense Evasion	6
48	AML.T0108 AI Agent	Command and Control	6
49	AML.T0096 AI Service API	Command and Control	5
50	AML.T0098 AI Agent Tool Credential Harvesting	Credential Access	5
51	AML.T0024 Exfiltration via AI Inference API	Exfiltration	4
52	AML.T0054 LLM Jailbreak	Privilege Escalation	4
53	AML.T0064 Gather RAG-Indexed Targets	Reconnaissance	4
54	AML.T0099 AI Agent Tool Data Poisoning	Persistence	4
55	AML.T0019 Publish Poisoned Datasets	Resource Development	3
56	AML.T0056 Extract LLM System Prompt	Exfiltration	3
57	AML.T0063 Discover AI Model Outputs	Discovery	3
58	AML.T0087 Gather Victim Identity Information	Reconnaissance	3
59	AML.T0010 AI Supply Chain Compromise	Initial Access	2
60	AML.T0014 Discover AI Model Family	Discovery	2
61	AML.T0015 Evade AI Model	Initial Access	2
62	AML.T0065 LLM Prompt Crafting	Resource Development	2
63	AML.T0100 AI Agent Clickbait	Execution	2
64	AML.T0102 Generate Malicious Commands	AI Attack Staging	2
65	AML.T0000 Search Open Technical Databases	Reconnaissance	1
66	AML.T0046 Spamming AI System with Chaff Data	Impact	1
67	AML.T0047 AI-Enabled Product or Service	AI Model Access	1
68	AML.T0066 Retrieval Content Crafting	Resource Development	1
69	AML.T0069 Discover LLM System Information	Discovery	1
70	AML.T0091 Use Alternate Authentication Material	Lateral Movement	1
71	AML.T0092 Manipulate User LLM Chat History	Defense Evasion	1
72	AML.T0002 Acquire Public AI Artifacts	Resource Development	0
73	AML.T0003 Search Victim-Owned Websites	Reconnaissance	0
74	AML.T0004 Search Application Repositories	Reconnaissance	0
75	AML.T0005 Create Proxy AI Model	AI Attack Staging	0
76	AML.T0008 Acquire Infrastructure	Resource Development	0
77	AML.T0013 Discover AI Model Ontology	Discovery	0
78	AML.T0016 Obtain Capabilities	Resource Development	0
79	AML.T0017 Develop Capabilities	Resource Development	0
80	AML.T0041 Physical Environment Access	AI Model Access	0
81	AML.T0042 Verify Attack	AI Attack Staging	0
82	AML.T0048 External Harms	Impact	0
83	AML.T0060 Publish Hallucinated Entities	Resource Development	0
84	AML.T0061 LLM Prompt Self-Replication	Persistence	0
85	AML.T0062 Discover LLM Hallucinations	Discovery	0
86	AML.T0067 LLM Trusted Output Components Manipulation	Defense Evasion	0
87	AML.T0068 LLM Prompt Obfuscation	Defense Evasion	0
88	AML.T0071 False RAG Entry Injection	Defense Evasion	0
89	AML.T0077 LLM Response Rendering	Exfiltration	0
90	AML.T0082 RAG Credential Harvesting	Credential Access	0
91	AML.T0088 Generate Deepfakes	AI Attack Staging	0
92	AML.T0089 Process Discovery	Discovery	0
93	AML.T0090 OS Credential Dumping	Credential Access	0
94	AML.T0094 Delay Execution of LLM Instructions	Defense Evasion	0
95	AML.T0095 Search Open Websites/Domains	Reconnaissance	0
96	AML.T0103 Deploy AI Agent	Execution	0
97	AML.T0104 Publish Poisoned AI Agent Tool	Resource Development	0

Track these techniques against your AI stack with real-time alerts.

Start 14-Day Free Trial