Tactics: AI Model Access, AI Attack Staging, Reconnaissance, Resource Development, Initial Access, Execution, Persistence, Defense Evasion, Discovery, Collection, Exfiltration, Impact, Privilege Escalation, Credential Access, Command and Control, Lateral Movement, Mitigations
[AML.TA0000] AI Model Access
- AML.T0040 AI Model Inference API Access
This node addresses MITRE ATLAS technique AML.T0043 (Craft Adversarial Perturbations for Evasion), aligning with EU AI Act Articles 15 and 9…
Bidda node: https://bidda.com/intelligence/mitre-atlas-craft-adversarial-perturbations
- AML.T0041 Physical Environment Access
This node addresses MITRE ATLAS technique AML.T0041 (Physical Environment Access). In addition to the attacks that take place purely in the digital domain, adversaries may also exploit the physical environment for their attacks…
Bidda node: https://bidda.com/intelligence/mitre-atlas-physical-environment-access
- AML.T0044 Full AI Model Access
This node addresses compliance obligations to prevent ML model fingerprinting under MITRE ATLAS technique AML.T0014, focusing on AI model identification prevention and adversarial profiling…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ml-model-fingerprinting
- AML.T0047 AI-Enabled Product or Service
This node addresses MITRE ATLAS technique AML.T0024.002, focusing on ML model stealing via inference API, aligning with EU AI Act Articles 9…
Bidda node: https://bidda.com/intelligence/mitre-atlas-model-stealing-inference-api
[AML.TA0001] AI Attack Staging
- AML.T0005 Create Proxy AI Model
This node addresses adversarial techniques under MITRE ATLAS AML.T0005 (Create Proxy Models), focusing on model replication and shadow model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-create-proxy-model
- AML.T0005.000 Train Proxy via Gathered AI Artifacts
This node addresses MITRE ATLAS technique AML.T0005.000 (Train Proxy via Gathered AI Artifacts). Proxy models may be trained from AI artifacts (such as data, model architectures, and pre-trained models) that are representative of the…
Bidda node: https://bidda.com/intelligence/mitre-atlas-train-proxy-via-gathered-ai-artifacts
- AML.T0005.001 Train Proxy via Replication
This node addresses MITRE ATLAS technique AML.T0005.001 (Train Proxy via Replication). Adversaries may replicate a private model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-train-proxy-via-replication
- AML.T0005.002 Use Pre-Trained Model
This node addresses MITRE ATLAS technique AML.T0005.002 (Use Pre-Trained Model). Adversaries may use an off-the-shelf pre-trained model as a proxy for the victim model to aid in staging the attack…
Bidda node: https://bidda.com/intelligence/mitre-atlas-use-pre-trained-model
- AML.T0018 Manipulate AI Model
This node addresses adversarial techniques under MITRE ATLAS AML.T0018 (Manipulate AI Model), focusing on poisoned models, modified architectures, and embedded…
Bidda node: https://bidda.com/intelligence/mitre-atlas-manipulate-ai-model
- AML.T0018.000 Poison AI Model
This node addresses MITRE ATLAS technique AML.T0018.000 (Poison AI Model). Adversaries may manipulate an AI model's weights to change it's behavior or performance, resulting in a poisoned model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-poison-ai-model
- AML.T0018.001 Modify AI Model Architecture
This node addresses MITRE ATLAS technique AML.T0018.001 (Modify AI Model Architecture). Adversaries may directly modify an AI model's architecture to re-define it's behavior…
Bidda node: https://bidda.com/intelligence/mitre-atlas-modify-ai-model-architecture
- AML.T0018.002 Embed Malware
This node addresses MITRE ATLAS technique AML.T0018.002 (Embed Malware). Adversaries may embed malicious code into AI Model files. AI models may be packaged as a combination of instructions and weights…
Bidda node: https://bidda.com/intelligence/mitre-atlas-embed-malware
- AML.T0042 Verify Attack
This node addresses MITRE ATLAS technique AML.T0042 (Verify Attack). Adversaries can verify the efficacy of their attack via an inference API or access to an offline copy of the target model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-verify-attack
- AML.T0043 Craft Adversarial Data
This node addresses MITRE ATLAS technique AML.T0015, focusing on adversarial evasion via explanation methods, aligns with EU AI Act Articles 9…
Bidda node: https://bidda.com/intelligence/mitre-atlas-evasion-via-explanation-methods
- AML.T0043.000 White-Box Optimization
This node addresses MITRE ATLAS technique AML.T0043.000 (White-Box Optimization). In White-Box Optimization, the adversary has full access to the target model and optimizes the adversarial example directly…
Bidda node: https://bidda.com/intelligence/mitre-atlas-white-box-optimization
- AML.T0043.001 Black-Box Optimization
This node addresses MITRE ATLAS technique AML.T0043.001 (Black-Box Optimization). In Black-Box attacks, the adversary has black-box (i.e. AI Model Inference API Access via API access) access to the target model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-black-box-optimization
- AML.T0043.002 Black-Box Transfer
This node addresses MITRE ATLAS technique AML.T0043.002 (Black-Box Transfer). In Black-Box Transfer attacks, the adversary uses one or more proxy models (trained via Create Proxy AI Model or Train Proxy via Replication) they have full…
Bidda node: https://bidda.com/intelligence/mitre-atlas-black-box-transfer
- AML.T0043.003 Manual Modification
This node addresses MITRE ATLAS technique AML.T0043.003 (Manual Modification). Adversaries may manually modify the input data to craft adversarial data…
Bidda node: https://bidda.com/intelligence/mitre-atlas-manual-modification
- AML.T0043.004 Insert Backdoor Trigger
This node addresses MITRE ATLAS technique AML.T0043.004 (Insert Backdoor Trigger). The adversary may add a perceptual trigger into inference data. The trigger may be imperceptible or non-obvious to humans…
Bidda node: https://bidda.com/intelligence/mitre-atlas-insert-backdoor-trigger
- AML.T0088 Generate Deepfakes
This node addresses MITRE ATLAS technique AML.T0088 (Generate Synthetic Media), focusing on adversarial AI tactics for creating deepfakes used in identity fraud…
Bidda node: https://bidda.com/intelligence/mitre-atlas-deepfake-generation
- AML.T0102 Generate Malicious Commands
This node addresses MITRE ATLAS technique AML.T0101 (Data Destruction via Compromised AI Agent), aligning with EU AI Act Articles 9, 14…
Bidda node: https://bidda.com/intelligence/mitre-atlas-data-destruction-agent
[AML.TA0002] Reconnaissance
- AML.T0000 Search Open Technical Databases
This node addresses MITRE ATLAS technique AML.T0000, focusing on adversarial reconnaissance via publicly available research, aligning with…
Bidda node: https://bidda.com/intelligence/mitre-atlas-reconnaissance-victim-research
- AML.T0000.000 Journals and Conference Proceedings
This node addresses MITRE ATLAS technique AML.T0000.000 (Journals and Conference Proceedings). Many of the publications accepted at premier artificial intelligence conferences and journals come from commercial labs…
Bidda node: https://bidda.com/intelligence/mitre-atlas-journals-and-conference-proceedings
- AML.T0000.001 Pre-Print Repositories
This node addresses MITRE ATLAS technique AML.T0000.001 (Pre-Print Repositories). Pre-Print repositories, such as arXiv, contain the latest academic research papers that haven't been peer reviewed…
Bidda node: https://bidda.com/intelligence/mitre-atlas-pre-print-repositories
- AML.T0000.002 Technical Blogs
This node addresses MITRE ATLAS technique AML.T0000.002 (Technical Blogs). Research labs at academic institutions and company R&D divisions often have blogs that highlight their use of artificial intelligence and its application to the…
Bidda node: https://bidda.com/intelligence/mitre-atlas-technical-blogs
- AML.T0001 Search Open AI Vulnerability Analysis
This node addresses compliance obligations under the EU AI Act for defending against MITRE ATLAS technique AML.T0001 (AI Reconnaissance via Adversarial Vulnerability Analysis), focusing on AI…
Bidda node: https://bidda.com/intelligence/mitre-atlas-adversarial-vulnerability-research
- AML.T0003 Search Victim-Owned Websites
This node addresses MITRE ATLAS technique AML.T0004, focusing on securing AI artifacts in public repositories under the EU…
Bidda node: https://bidda.com/intelligence/mitre-atlas-application-repository-search
- AML.T0004 Search Application Repositories
This node addresses MITRE ATLAS technique AML.T0004 (Search Application Repositories). Adversaries may search open application repositories during targeting…
Bidda node: https://bidda.com/intelligence/mitre-atlas-search-application-repositories
- AML.T0006 Active Scanning
This node addresses MITRE ATLAS technique AML.T0002, focusing on adversaries acquiring public ML artifacts for attack…
Bidda node: https://bidda.com/intelligence/mitre-atlas-acquire-public-ml-artifacts
- AML.T0064 Gather RAG-Indexed Targets
This node addresses MITRE ATLAS technique AML.T0064 (Gather RAG-Indexed Targets). Adversaries may identify data sources used in retrieval augmented generation (RAG) systems for targeting purposes…
Bidda node: https://bidda.com/intelligence/mitre-atlas-gather-rag-indexed-targets
- AML.T0087 Gather Victim Identity Information
This node addresses MITRE ATLAS technique AML.T0087 (Gather Victim Identity Information). Adversaries may gather information about the victim's identity that can be used during targeting…
Bidda node: https://bidda.com/intelligence/mitre-atlas-gather-victim-identity-information
- AML.T0095 Search Open Websites/Domains
This node addresses MITRE ATLAS technique AML.T0095 (Search Open Websites/Domains). Adversaries may search public websites and/or domains for information about victims that can be used during targeting…
Bidda node: https://bidda.com/intelligence/mitre-atlas-search-open-websites-domains
- AML.T0095.000 Code Repositories
This node addresses MITRE ATLAS technique AML.T0095.000 (Code Repositories). Adversaries may search public code repositories for information about a victim or victim system that can be used during targeting…
Bidda node: https://bidda.com/intelligence/mitre-atlas-code-repositories
[AML.TA0003] Resource Development
- AML.T0002 Acquire Public AI Artifacts
This node addresses MITRE ATLAS technique AML.T0003, focusing on adversary reconnaissance via victim-owned websites and APIs…
Bidda node: https://bidda.com/intelligence/mitre-atlas-victim-website-reconnaissance
- AML.T0002.000 Datasets
This node addresses MITRE ATLAS technique AML.T0002.000 (Datasets). Adversaries may collect public datasets to use in their operations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-datasets
- AML.T0002.001 Models
This node addresses MITRE ATLAS technique AML.T0002.001 (Models). Adversaries may acquire public models to use in their operations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-models
- AML.T0002.002 AI Agent Configuration
This node addresses MITRE ATLAS technique AML.T0002.002 (AI Agent Configuration). Adversaries may acquire publicly accessible AI agent configuration files to understand agent capabilities, gain unauthorized access to tools and data…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-agent-configuration
- AML.T0008 Acquire Infrastructure
This node addresses compliance obligations under the EU AI Act for defending against MITRE ATLAS technique AML.T0010 (AI Supply Chain Compromise), specifically requiring pre-trained model integrity…
Bidda node: https://bidda.com/intelligence/mitre-atlas-acquire-compromised-ml-artifact
- AML.T0008.000 AI Development Workspaces
This node addresses MITRE ATLAS technique AML.T0008.000 (AI Development Workspaces). Developing and staging AI attacks often requires expensive compute resources…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-development-workspaces
- AML.T0008.001 Consumer Hardware
This node addresses MITRE ATLAS technique AML.T0008.001 (Consumer Hardware). Adversaries may acquire consumer hardware to conduct their attacks. Owning the hardware provides the adversary with complete control of the environment…
Bidda node: https://bidda.com/intelligence/mitre-atlas-t0008-001-commodity-grade-hardware-risk
- AML.T0008.002 Domains
This node addresses MITRE ATLAS technique AML.T0008.002 (Domains). Adversaries may acquire domains that can be used during targeting. Domain names are the human readable names used to represent one or more IP addresses…
Bidda node: https://bidda.com/intelligence/mitre-atlas-domains
- AML.T0008.003 Physical Countermeasures
This node addresses MITRE ATLAS technique AML.T0008.003 (Physical Countermeasures). Adversaries may acquire or manufacture physical countermeasures to aid or support their attack…
Bidda node: https://bidda.com/intelligence/mitre-atlas-physical-countermeasures
- AML.T0008.004 Serverless
This node addresses MITRE ATLAS technique AML.T0008.004 (Serverless). Adversaries may purchase and configure serverless cloud infrastructure, such as Cloudflare Workers, AWS Lambda functions, or Google Apps Scripts, that can be used…
Bidda node: https://bidda.com/intelligence/mitre-atlas-serverless
- AML.T0008.005 AI Service Proxies
This node addresses MITRE ATLAS technique AML.T0008.005 (AI Service Proxies). Adversaries may utilize commercial proxy services that resell access to AI services such as frontier model APIs…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-service-proxies
- AML.T0016 Obtain Capabilities
This node addresses compliance obligations under MITRE ATLAS technique AML.T0016 for securing AI/ML capabilities against adversarial attacks, aligning with EU AI Act…
Bidda node: https://bidda.com/intelligence/mitre-atlas-obtain-ml-capabilities
- AML.T0016.000 Adversarial AI Attack Implementations
This node addresses MITRE ATLAS technique AML.T0016.000 (Adversarial AI Attack Implementations). Adversaries may search for existing open source implementations of AI attacks…
Bidda node: https://bidda.com/intelligence/mitre-atlas-adversarial-ai-attack-implementations
- AML.T0016.001 Software Tools
This node addresses MITRE ATLAS technique AML.T0016.001 (Software Tools). Adversaries may search for and obtain software tools to support their operations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-software-tools
- AML.T0016.002 Generative AI
This node addresses MITRE ATLAS technique AML.T0016.002 (Generative AI). Adversaries may search for and obtain generative AI models or tools, such as large language models (LLMs), to assist them in various steps of their operation…
Bidda node: https://bidda.com/intelligence/mitre-atlas-generative-ai
- AML.T0017 Develop Capabilities
This node addresses compliance obligations under MITRE ATLAS technique AML.T0017 for developing adversarial AI capabilities and attack…
Bidda node: https://bidda.com/intelligence/mitre-atlas-develop-adversarial-capabilities
- AML.T0017.000 Adversarial AI Attacks
This node addresses MITRE ATLAS technique AML.T0017.000 (Adversarial AI Attacks). Adversaries may develop their own adversarial attacks. They may leverage existing libraries as a starting point (Adversarial AI Attack Implementations)…
Bidda node: https://bidda.com/intelligence/mitre-atlas-adversarial-ai-attacks
- AML.T0019 Publish Poisoned Datasets
This node addresses MITRE ATLAS technique AML.T0019 (Publish Poisoned Datasets). Adversaries may Poison Training Data and publish it to a public location…
Bidda node: https://bidda.com/intelligence/mitre-atlas-publish-poisoned-datasets
- AML.T0020 Poison Training Data
This node addresses MITRE ATLAS technique AML.T0020 (Training Data Poisoning), focusing on adversarial corruption of training and fine-tuning datasets to embed…
Bidda node: https://bidda.com/intelligence/mitre-atlas-data-poisoning
- AML.T0021 Establish Accounts
This node addresses MITRE ATLAS technique AML.T0021 (Establish Accounts). Adversaries may create accounts with various services for use in targeting, to gain access to resources needed in AI Attack Staging, or for victim impersonation…
Bidda node: https://bidda.com/intelligence/mitre-atlas-establish-accounts
- AML.T0058 Publish Poisoned Models
This node addresses MITRE ATLAS technique AML.T0058 (Publish Poisoned Models). Adversaries may publish a poisoned model to a public location such as a model registry or code repository…
Bidda node: https://bidda.com/intelligence/mitre-atlas-publish-poisoned-models
- AML.T0060 Publish Hallucinated Entities
This node addresses MITRE ATLAS technique AML.T0060 (Publish Hallucinated Entities). Adversaries may create an entity they control, such as a software package, website, or email address to a source hallucinated by an LLM…
Bidda node: https://bidda.com/intelligence/mitre-atlas-publish-hallucinated-entities
- AML.T0065 LLM Prompt Crafting
This node addresses MITRE ATLAS technique AML.T0065 (LLM Prompt Crafting). Adversaries may use their acquired knowledge of the target generative AI system to craft prompts that bypass its defenses and allow malicious instructions to be…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-prompt-crafting
- AML.T0066 Retrieval Content Crafting
This node addresses compliance obligations to protect against MITRE ATLAS Membership Inference Attacks (AML.T0024.000) on ML training data, aligning with EU AI Act Articles…
Bidda node: https://bidda.com/intelligence/mitre-atlas-membership-inference-attack
- AML.T0079 Stage Capabilities
This node addresses MITRE ATLAS technique AML.T0079 (Stage Capabilities). Adversaries may upload, install, or otherwise set up capabilities that can be used during targeting…
Bidda node: https://bidda.com/intelligence/mitre-atlas-stage-capabilities
- AML.T0104 Publish Poisoned AI Agent Tool
This node addresses the MITRE ATLAS technique AML.T0053, focusing on preventing code execution via hijacked AI agents, aligned with EU…
Bidda node: https://bidda.com/intelligence/mitre-atlas-execution-via-agent
[AML.TA0004] Initial Access
- AML.T0010 AI Supply Chain Compromise
This node addresses adversarial techniques under MITRE ATLAS AML.T0010, focusing on AI supply chain compromises such as targeting model repositories, pre-trained weights, datasets…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-supply-chain-compromise
- AML.T0010.000 Hardware
This node addresses MITRE ATLAS technique AML.T0010.000 (Hardware). Adversaries may target AI systems by disrupting or manipulating the hardware supply chain…
Bidda node: https://bidda.com/intelligence/mitre-atlas-hardware
- AML.T0010.001 AI Software
This node addresses MITRE ATLAS technique AML.T0010.001 (AI Software). Adversaries may target software packages that are commonly used in AI-enabled systems or are part of the AI DevOps lifecycle…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-software
- AML.T0010.002 Data
This node addresses MITRE ATLAS technique AML.T0010.002 (Data). Data is a key vector of supply chain compromise for adversaries. Every AI project will require some form of data…
Bidda node: https://bidda.com/intelligence/mitre-atlas-data
- AML.T0010.003 Model
This node addresses MITRE ATLAS technique AML.T0010.003 (Model). AI-enabled systems often rely on open sourced models in various ways. Most commonly, the victim organization may be using these models for fine tuning…
Bidda node: https://bidda.com/intelligence/mitre-atlas-model
- AML.T0010.004 Container Registry
This node addresses MITRE ATLAS technique AML.T0010.004 (Container Registry). An adversary may compromise a victim's container registry by pushing a manipulated container image and overwriting an existing container name and/or tag…
Bidda node: https://bidda.com/intelligence/mitre-atlas-container-registry
- AML.T0010.005 AI Agent Tool
This node addresses MITRE ATLAS technique AML.T0010.005 (AI Agent Tool). Adversaries may target AI agent tools as a means to compromise a victim's AI supply chain…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-agent-tool
- AML.T0012 Valid Accounts
This node addresses compliance obligations to mitigate MITRE ATLAS technique AML.T0012 (Valid ML Platform Account Abuse) for unauthorised AI…
Bidda node: https://bidda.com/intelligence/mitre-atlas-valid-ml-platform-accounts
- AML.T0015 Evade AI Model
This node addresses MITRE ATLAS technique AML.T0015 (Evade AI Model). Adversaries can Craft Adversarial Data that prevents an AI model from correctly identifying the contents of the data or Generate Deepfakes that fools an AI model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-evade-ai-model
- AML.T0049 Exploit Public-Facing Application
This node addresses MITRE ATLAS technique AML.T0049 (Exploit Public-Facing Application). Adversaries may attempt to take advantage of a weakness in an Internet-facing computer or program using software, data, or commands in order to cause…
Bidda node: https://bidda.com/intelligence/mitre-atlas-exploit-public-facing-application
- AML.T0052 Phishing
This node addresses MITRE ATLAS technique AML.T0052, focusing on spearphishing attacks targeting ML researchers, with compliance obligations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-phishing-ml-researcher
- AML.T0052.000 Spearphishing via Social Engineering LLM
This node addresses MITRE ATLAS technique AML.T0052.000 (Spearphishing via Social Engineering LLM). Adversaries may turn LLMs into targeted social engineers. LLMs are capable of interacting with users via text conversations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-spearphishing-via-social-engineering-llm
- AML.T0052.001 Deepfake-Assisted Phishing
This node addresses MITRE ATLAS technique AML.T0052.001 (Deepfake-Assisted Phishing). Adversaries may use deepfakes (AI-generated synthetic images, audio, or video) in phishing campaigns to impersonate trusted individuals, executives, or…
Bidda node: https://bidda.com/intelligence/mitre-atlas-deepfake-assisted-phishing
- AML.T0078 Drive-by Compromise
This node addresses MITRE ATLAS technique AML.T0078 (Drive-by Compromise). Adversaries may gain access to an AI system through a user visiting a website over the normal course of browsing, or an AI agent retrieving information from the…
Bidda node: https://bidda.com/intelligence/mitre-atlas-drive-by-compromise
- AML.T0093 Prompt Infiltration via Public-Facing Application
This node addresses MITRE ATLAS technique AML.T0093 (Prompt Infiltration via Public-Facing Application). An adversary may introduce malicious prompts into the victim's system via a public-facing application with the intention of it being…
Bidda node: https://bidda.com/intelligence/mitre-atlas-prompt-infiltration-via-public-facing-application
[AML.TA0005] Execution
- AML.T0011 User Execution
This node addresses compliance obligations to prevent user execution of malicious ML models and adversarial notebooks under MITRE ATLAS technique AML.T0011, aligning with EU AI Act Article 15 (accuracy, robustness and cybersecurity)…
Bidda node: https://bidda.com/intelligence/mitre-atlas-user-execution-ml-model
- AML.T0011.000 Unsafe AI Artifacts
This node addresses MITRE ATLAS technique AML.T0011.000 (Unsafe AI Artifacts). Adversaries may develop unsafe AI artifacts that when executed have a deleterious effect…
Bidda node: https://bidda.com/intelligence/mitre-atlas-unsafe-ai-artifacts
- AML.T0011.001 Malicious Package
This node addresses MITRE ATLAS technique AML.T0011.001 (Malicious Package). Adversaries may develop malicious software packages that when imported by a user have a deleterious effect. Malicious packages may behave as expected to the user…
Bidda node: https://bidda.com/intelligence/mitre-atlas-malicious-package
- AML.T0011.002 Poisoned AI Agent Tool
This node addresses MITRE ATLAS technique AML.T0011.002 (Poisoned AI Agent Tool). A victim may invoke a poisoned tool when interacting with their AI agent…
Bidda node: https://bidda.com/intelligence/mitre-atlas-poisoned-ai-agent-tool
- AML.T0011.003 Malicious Link
This node addresses MITRE ATLAS technique AML.T0011.003 (Malicious Link). An adversary may rely upon a user clicking a malicious link in order to gain execution…
Bidda node: https://bidda.com/intelligence/mitre-atlas-malicious-link
- AML.T0050 Command and Scripting Interpreter
This node addresses MITRE ATLAS technique AML.T0007, focusing on defending against OSINT-based discovery of AI artifacts, aligned with EU AI Act Articles 9…
Bidda node: https://bidda.com/intelligence/mitre-atlas-search-victim-ai-artifacts
- AML.T0051 LLM Prompt Injection
This node addresses MITRE ATLAS technique AML.T0051 (LLM Prompt Injection), focusing on adversarial inputs designed to manipulate large language model behavior and bypass safety…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-prompt-injection
- AML.T0051.000 Direct
This node addresses MITRE ATLAS technique AML.T0051.000 (Direct). An adversary may inject prompts directly as a user of the LLM…
Bidda node: https://bidda.com/intelligence/mitre-atlas-direct
- AML.T0051.001 Indirect
This node addresses MITRE ATLAS technique AML.T0051.001 (Indirect). An adversary may inject prompts indirectly via separate data channel ingested by the LLM such as include text or multimedia pulled from databases or websites…
Bidda node: https://bidda.com/intelligence/mitre-atlas-indirect
- AML.T0051.002 Triggered
This node addresses MITRE ATLAS technique AML.T0051.002 (Triggered). An adversary may trigger a prompt injection via a user action or event that occurs within the victim's environment…
Bidda node: https://bidda.com/intelligence/mitre-atlas-triggered
- AML.T0053 AI Agent Tool Invocation
This node addresses MITRE ATLAS technique AML.T0053 (AI Agent Tool Invocation). Adversaries may use their access to an AI agent to invoke tools the agent has access to…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-agent-tool-invocation
- AML.T0100 AI Agent Clickbait
This node addresses MITRE ATLAS tactic AML.TA0012, focusing on privilege escalation via AI agent tool access, aligning with EU AI Act…
Bidda node: https://bidda.com/intelligence/mitre-atlas-privilege-escalation-agent
- AML.T0103 Deploy AI Agent
This node addresses MITRE ATLAS technique AML.T0086, focusing on data exfiltration via compromised AI agent tool calls, aligning with EU AI Act…
Bidda node: https://bidda.com/intelligence/mitre-atlas-exfiltration-agent
[AML.TA0006] Persistence
- AML.T0018 Manipulate AI Model
This node addresses adversarial techniques under MITRE ATLAS AML.T0018 (Manipulate AI Model), focusing on poisoned models, modified architectures, and embedded…
Bidda node: https://bidda.com/intelligence/mitre-atlas-manipulate-ai-model
- AML.T0018.000 Poison AI Model
This node addresses MITRE ATLAS technique AML.T0018.000 (Poison AI Model). Adversaries may manipulate an AI model's weights to change it's behavior or performance, resulting in a poisoned model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-poison-ai-model
- AML.T0018.001 Modify AI Model Architecture
This node addresses MITRE ATLAS technique AML.T0018.001 (Modify AI Model Architecture). Adversaries may directly modify an AI model's architecture to re-define it's behavior…
Bidda node: https://bidda.com/intelligence/mitre-atlas-modify-ai-model-architecture
- AML.T0018.002 Embed Malware
This node addresses MITRE ATLAS technique AML.T0018.002 (Embed Malware). Adversaries may embed malicious code into AI Model files. AI models may be packaged as a combination of instructions and weights…
Bidda node: https://bidda.com/intelligence/mitre-atlas-embed-malware
- AML.T0020 Poison Training Data
This node addresses MITRE ATLAS technique AML.T0020 (Training Data Poisoning), focusing on adversarial corruption of training and fine-tuning datasets to embed…
Bidda node: https://bidda.com/intelligence/mitre-atlas-data-poisoning
- AML.T0061 LLM Prompt Self-Replication
This node addresses MITRE ATLAS technique AML.T0061 (LLM Prompt Self-Replication). An adversary may use a carefully crafted LLM Prompt Injection designed to cause the LLM to replicate the prompt as part of its output…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-prompt-self-replication
- AML.T0070 RAG Poisoning
This node addresses MITRE ATLAS RAG Poisoning Techniques (AML.T0070), focusing on adversarial attacks that compromise retrieval-augmented generation systems by poisoning…
Bidda node: https://bidda.com/intelligence/mitre-atlas-rag-poisoning
- AML.T0080 AI Agent Context Poisoning
This node addresses MITRE ATLAS technique AML.T0080 (AI Agent Context Poisoning). Adversaries may attempt to manipulate the context used by an AI agent's large language model (LLM) to influence the responses it generates or actions it…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-agent-context-poisoning
- AML.T0080.000 Memory
This node addresses MITRE ATLAS technique AML.T0080.000 (Memory). Adversaries may manipulate the memory of a large language model (LLM) in order to persist changes to the LLM to future chat sessions…
Bidda node: https://bidda.com/intelligence/mitre-atlas-memory
- AML.T0080.001 Thread
This node addresses MITRE ATLAS technique AML.T0080.001 (Thread). Adversaries may introduce malicious instructions into a chat thread of a large language model (LLM) to cause behavior changes which persist for the remainder of the thread…
Bidda node: https://bidda.com/intelligence/mitre-atlas-thread
- AML.T0081 Modify AI Agent Configuration
This node addresses MITRE ATLAS technique AML.T0081 (Modify AI Agent Configuration). Adversaries may modify the configuration files for AI agents on a system…
Bidda node: https://bidda.com/intelligence/mitre-atlas-modify-ai-agent-configuration
- AML.T0093 Prompt Infiltration via Public-Facing Application
This node addresses MITRE ATLAS technique AML.T0093 (Prompt Infiltration via Public-Facing Application). An adversary may introduce malicious prompts into the victim's system via a public-facing application with the intention of it being…
Bidda node: https://bidda.com/intelligence/mitre-atlas-prompt-infiltration-via-public-facing-application
- AML.T0099 AI Agent Tool Data Poisoning
This node addresses MITRE ATLAS tactic AML.TA0015, focusing on lateral movement via compromised AI agents, aligning with EU AI Act Articles 9…
Bidda node: https://bidda.com/intelligence/mitre-atlas-lateral-movement-agent
- AML.T0110 AI Agent Tool Poisoning
This node addresses compliance obligations under the EU AI Act for mitigating adversarial AI techniques like MITRE ATLAS AML.TA0003 (Resource Development via AI…
Bidda node: https://bidda.com/intelligence/mitre-atlas-resource-development-agent
[AML.TA0007] Defense Evasion
- AML.T0015 Evade AI Model
This node addresses MITRE ATLAS technique AML.T0015 (Evade AI Model). Adversaries can Craft Adversarial Data that prevents an AI model from correctly identifying the contents of the data or Generate Deepfakes that fools an AI model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-evade-ai-model
- AML.T0054 LLM Jailbreak
This node addresses MITRE ATLAS technique AML.T0054 (LLM Jailbreak), focusing on adversarial prompt engineering to bypass AI safety controls, with…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-jailbreak
- AML.T0067 LLM Trusted Output Components Manipulation
This node addresses MITRE ATLAS technique AML.T0067 (LLM Trusted Output Components Manipulation). Adversaries may utilize prompts to a large language model (LLM) which manipulate various components of its response in order to make it…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-trusted-output-components-manipulation
- AML.T0067.000 Citations
This node addresses MITRE ATLAS technique AML.T0067.000 (Citations). Adversaries may manipulate the citations provided in an AI system's response, in order to make it appear trustworthy…
Bidda node: https://bidda.com/intelligence/mitre-atlas-citations
- AML.T0068 LLM Prompt Obfuscation
This node addresses MITRE ATLAS technique AML.T0068 (LLM Prompt Obfuscation). Adversaries may hide or otherwise obfuscate prompt injections or retrieval content to avoid detection from humans, large language model (LLM) guardrails, or…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-prompt-obfuscation
- AML.T0071 False RAG Entry Injection
This node addresses MITRE ATLAS technique AML.T0071, focusing on False RAG Entry Injection for defense evasion through misinformation insertion in…
Bidda node: https://bidda.com/intelligence/mitre-atlas-false-rag-injection
- AML.T0073 Impersonation
This node addresses MITRE ATLAS technique AML.T0073 (Impersonation). Adversaries may impersonate a trusted person or organization in order to persuade and trick a target into performing some action on their behalf…
Bidda node: https://bidda.com/intelligence/mitre-atlas-impersonation
- AML.T0074 Masquerading
This node addresses MITRE ATLAS technique AML.T0074 (Masquerading). Adversaries may attempt to manipulate features of their artifacts to make them appear legitimate or benign to users and/or security tools…
Bidda node: https://bidda.com/intelligence/mitre-atlas-masquerading
- AML.T0076 Corrupt AI Model
This node addresses MITRE ATLAS technique AML.T0076 (Corrupt AI Model). An adversary may purposefully corrupt a malicious AI model file so that it cannot be successfully deserialized in order to evade detection by a model scanner…
Bidda node: https://bidda.com/intelligence/mitre-atlas-corrupt-ai-model
- AML.T0081 Modify AI Agent Configuration
This node addresses MITRE ATLAS technique AML.T0081 (Modify AI Agent Configuration). Adversaries may modify the configuration files for AI agents on a system…
Bidda node: https://bidda.com/intelligence/mitre-atlas-modify-ai-agent-configuration
- AML.T0092 Manipulate User LLM Chat History
This node addresses MITRE ATLAS technique AML.T0092 (Manipulate User LLM Chat History). Adversaries may manipulate a user's large language model (LLM) chat history to cover the tracks of their malicious behavior…
Bidda node: https://bidda.com/intelligence/mitre-atlas-manipulate-user-llm-chat-history
- AML.T0094 Delay Execution of LLM Instructions
This node addresses MITRE ATLAS technique AML.T0094 (Delay Execution of LLM Instructions). Adversaries may include instructions to be followed by the AI system in response to a future event, such as a specific keyword or the next…
Bidda node: https://bidda.com/intelligence/mitre-atlas-delay-execution-of-llm-instructions
- AML.T0097 Virtualization/Sandbox Evasion
This node addresses MITRE ATLAS technique AML.T0097 (Virtualization/Sandbox Evasion). Adversaries may employ various means to detect and avoid virtualization and analysis environments…
Bidda node: https://bidda.com/intelligence/mitre-atlas-virtualization-sandbox-evasion
- AML.T0107 Exploitation for Defense Evasion
This node addresses MITRE ATLAS technique AML.T0107 (Exploitation for Defense Evasion). Adversaries may exploit a system or application vulnerability to bypass security features…
Bidda node: https://bidda.com/intelligence/mitre-atlas-exploitation-for-defense-evasion
- AML.T0109 AI Supply Chain Rug Pull
This node addresses MITRE ATLAS technique AML.T0109 (AI Supply Chain Rug Pull). Adversaries may publish legitimate AI components or software, gain user adoption, then push an update with a malicious variant, leading to AI Supply Chain…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-supply-chain-rug-pull
- AML.T0111 AI Supply Chain Reputation Inflation
This node addresses MITRE ATLAS technique AML.T0111 (AI Supply Chain Reputation Inflation). AI Supply Chain Reputation Inflation is the process of building or leveraging genuinely credible-looking trust signals to increase the perceived…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-supply-chain-reputation-inflation
[AML.TA0008] Discovery
- AML.T0007 Discover AI Artifacts
This node addresses MITRE ATLAS technique AML.T0007 (Discover AI Artifacts). Adversaries may search private sources to identify AI learning artifacts that exist on the system and gather information about them…
Bidda node: https://bidda.com/intelligence/mitre-atlas-discover-ai-artifacts
- AML.T0013 Discover AI Model Ontology
This node addresses MITRE ATLAS technique AML.T0013 (Discover AI Model Ontology). Adversaries may discover the ontology of an AI model's output space, for example, the types of objects a model can detect…
Bidda node: https://bidda.com/intelligence/mitre-atlas-discover-ai-model-ontology
- AML.T0014 Discover AI Model Family
This node addresses MITRE ATLAS technique AML.T0043 (Craft Adversarial Data), aligning with EU AI Act Articles 9 and…
Bidda node: https://bidda.com/intelligence/mitre-atlas-craft-adversarial-data
- AML.T0062 Discover LLM Hallucinations
This node addresses MITRE ATLAS technique AML.T0062 (Discover LLM Hallucinations). Adversaries may prompt large language models and identify hallucinated entities…
Bidda node: https://bidda.com/intelligence/mitre-atlas-discover-llm-hallucinations
- AML.T0063 Discover AI Model Outputs
This node addresses MITRE ATLAS technique AML.T0063 (Discover AI Model Outputs). Adversaries may discover model outputs, such as class scores, whose presence is not required for the system to function and are not intended for use by the…
Bidda node: https://bidda.com/intelligence/mitre-atlas-discover-ai-model-outputs
- AML.T0069 Discover LLM System Information
This node addresses MITRE ATLAS technique AML.T0069 (Discover LLM System Information). The adversary is trying to discover something about the large language model's (LLM) system information…
Bidda node: https://bidda.com/intelligence/mitre-atlas-discover-llm-system-information
- AML.T0069.000 Special Character Sets
This node addresses MITRE ATLAS technique AML.T0069.000 (Special Character Sets). Adversaries may discover delimiters and special characters sets used by the large language model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-special-character-sets
- AML.T0069.001 System Instruction Keywords
This node addresses MITRE ATLAS technique AML.T0069.001 (System Instruction Keywords). Adversaries may discover keywords that have special meaning to the large language model (LLM), such as function names or object names…
Bidda node: https://bidda.com/intelligence/mitre-atlas-system-instruction-keywords
- AML.T0069.002 System Prompt
This node addresses MITRE ATLAS technique AML.T0069.002 (System Prompt). Adversaries may discover a large language model's system instructions provided by the AI system builder to learn about the system's capabilities and circumvent its…
Bidda node: https://bidda.com/intelligence/mitre-atlas-system-prompt
- AML.T0075 Cloud Service Discovery
This node addresses MITRE ATLAS technique AML.T0075 (Cloud Service Discovery). Adversaries may attempt to enumerate the cloud services running on a system after gaining access…
Bidda node: https://bidda.com/intelligence/mitre-atlas-cloud-service-discovery
- AML.T0084 Discover AI Agent Configuration
This node addresses MITRE ATLAS technique AML.T0084 (Discover AI Agent Configuration). Adversaries may attempt to discover configuration information for AI agents present on the victim's system…
Bidda node: https://bidda.com/intelligence/mitre-atlas-discover-ai-agent-configuration
- AML.T0084.000 Embedded Knowledge
This node addresses MITRE ATLAS technique AML.T0084.000 (Embedded Knowledge). Adversaries may attempt to discover the data sources a particular agent can access. The AI agent's configuration may reveal data sources or knowledge…
Bidda node: https://bidda.com/intelligence/mitre-atlas-embedded-knowledge
- AML.T0084.001 Tool Definitions
This node addresses MITRE ATLAS technique AML.T0084.001 (Tool Definitions). Adversaries may discover the tools the AI agent has access to…
Bidda node: https://bidda.com/intelligence/mitre-atlas-tool-definitions
- AML.T0084.002 Activation Triggers
This node addresses MITRE ATLAS technique AML.T0084.002 (Activation Triggers). Adversaries may discover keywords or other triggers (such as incoming emails, documents being added, incoming message, or other workflows) that activate an…
Bidda node: https://bidda.com/intelligence/mitre-atlas-activation-triggers
- AML.T0084.003 Call Chains
This node addresses MITRE ATLAS technique AML.T0084.003 (Call Chains). Adversaries may extract call chains from AI agent configurations, which can reveal potentially targets for remote code execution (RCE) or other vulnerabilities…
Bidda node: https://bidda.com/intelligence/mitre-atlas-call-chains
- AML.T0089 Process Discovery
This node addresses MITRE ATLAS technique AML.T0073, focusing on identity theft via AI synthetic media, aligning with EU AI Act Articles…
Bidda node: https://bidda.com/intelligence/mitre-atlas-identity-theft-synthetic-media
[AML.TA0009] Collection
- AML.T0035 AI Artifact Collection
This node addresses adversarial techniques under MITRE ATLAS AML.T0035 (AI Artifact Collection), focusing on the theft of AI model weights, training datasets, and…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-artifact-collection
- AML.T0036 Data from Information Repositories
This node addresses MITRE ATLAS technique AML.T0036 (Data from Information Repositories). Adversaries may leverage information repositories to mine valuable information…
Bidda node: https://bidda.com/intelligence/mitre-atlas-data-from-information-repositories
- AML.T0037 Data from Local System
This node addresses MITRE ATLAS technique AML.T0037 (Data from Local System). Adversaries may search local system sources, such as file systems and configuration files or local databases, to find files of interest and sensitive data prior…
Bidda node: https://bidda.com/intelligence/mitre-atlas-data-from-local-system
- AML.T0085 Data from AI Services
This node addresses MITRE ATLAS technique AML.T0085 (Data from AI Services). Adversaries may use their access to a victim organization's AI-enabled services to collect proprietary or otherwise sensitive information…
Bidda node: https://bidda.com/intelligence/mitre-atlas-data-from-ai-services
- AML.T0085.000 RAG Databases
This node addresses MITRE ATLAS technique AML.T0085.000 (RAG Databases). Adversaries may prompt the AI service to retrieve data from a RAG database. This can include the majority of an organization's internal documents…
Bidda node: https://bidda.com/intelligence/mitre-atlas-rag-databases
- AML.T0085.001 AI Agent Tools
This node addresses MITRE ATLAS technique AML.T0085.001 (AI Agent Tools). Adversaries may prompt the AI service to invoke various tools the agent has access to. Tools may retrieve data from different APIs or services in an organization…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-agent-tools
[AML.TA0010] Exfiltration
- AML.T0024 Exfiltration via AI Inference API
This node addresses adversarial AI techniques under MITRE ATLAS AML.T0024, including model inversion, training data membership inference, and AI model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-exfiltration-via-ai-inference
- AML.T0024.000 Infer Training Data Membership
This node addresses MITRE ATLAS technique AML.T0024.000 (Infer Training Data Membership). Adversaries may infer the membership of a data sample or global characteristics of the data in its training set, which raises privacy concerns…
Bidda node: https://bidda.com/intelligence/mitre-atlas-infer-training-data-membership
- AML.T0024.001 Invert AI Model
This node addresses MITRE ATLAS technique AML.T0024.001 (Invert AI Model). AI models' training data could be reconstructed by exploiting the confidence scores that are available via an inference API…
Bidda node: https://bidda.com/intelligence/mitre-atlas-invert-ai-model
- AML.T0024.002 Extract AI Model
This node addresses MITRE ATLAS technique AML.T0024.002 (Extract AI Model). Adversaries may extract a functional copy of a private model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-extract-ai-model
- AML.T0025 Exfiltration via Cyber Means
This node addresses MITRE ATLAS technique AML.T0020 (Manipulate Training Data) under the EU AI Act Articles 9, 10, and 15, focusing on implementing…
Bidda node: https://bidda.com/intelligence/mitre-atlas-manipulate-training-data
- AML.T0056 Extract LLM System Prompt
This node addresses adversarial techniques for extracting system prompts from large language models as defined in MITRE ATLAS AML.T0056 (LLM System Prompt Extraction), focusing on compliance obligations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-extract-system-prompt
- AML.T0057 LLM Data Leakage
This node addresses MITRE ATLAS technique AML.T0057 (LLM Data Leakage), focusing on adversarial exfiltration of training data, system prompts, and confidential information…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-data-leakage
- AML.T0077 LLM Response Rendering
This node addresses MITRE ATLAS technique AML.T0077 (LLM Response Rendering). An adversary may get a large language model (LLM) to respond with private information that is hidden from the user when the response is rendered by the user's…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-response-rendering
- AML.T0086 Exfiltration via AI Agent Tool Invocation
This node addresses MITRE ATLAS technique AML.T0086 (Exfiltration via AI Agent Tool Invocation). AI agent tools capable of performing write operations may be invoked to exfiltrate data to an adversary…
Bidda node: https://bidda.com/intelligence/mitre-atlas-exfiltration-via-ai-agent-tool-invocation
[AML.TA0011] Impact
- AML.T0015 Evade AI Model
This node addresses MITRE ATLAS technique AML.T0015 (Evade AI Model). Adversaries can Craft Adversarial Data that prevents an AI model from correctly identifying the contents of the data or Generate Deepfakes that fools an AI model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-evade-ai-model
- AML.T0029 Denial of AI Service
This node addresses adversarial techniques under MITRE ATLAS AML.T0029 (Denial of AI Service), focusing on resource exhaustion and model overload attacks…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-denial-of-service
- AML.T0031 Erode AI Model Integrity
This node addresses MITRE ATLAS technique AML.T0031 (Erode AI Model Integrity). Adversaries may degrade the target model's performance with adversarial data inputs to erode confidence in the system over time…
Bidda node: https://bidda.com/intelligence/mitre-atlas-erode-ai-model-integrity
- AML.T0034 Cost Harvesting
This node addresses adversarial AI cost harvesting techniques under MITRE ATLAS AML.T0034, including Excessive Query, Resource-Intensive Prompt, and Agentic…
Bidda node: https://bidda.com/intelligence/mitre-atlas-cost-harvesting
- AML.T0034.000 Excessive Queries
This node addresses MITRE ATLAS technique AML.T0034.000 (Excessive Queries). Adversaries may send an excessive number of otherwise normal or low-complexity queries to an AI system with the goal of overwhelming its capacity and increasing…
Bidda node: https://bidda.com/intelligence/mitre-atlas-excessive-queries
- AML.T0034.001 Resource-Intensive Queries
This node addresses MITRE ATLAS technique AML.T0034.001 (Resource-Intensive Queries). Adversaries may craft inputs specifically designed to increase the compute resources required for processing…
Bidda node: https://bidda.com/intelligence/mitre-atlas-resource-intensive-queries
- AML.T0034.002 Agentic Resource Consumption
This node addresses MITRE ATLAS technique AML.T0034.002 (Agentic Resource Consumption). Adversaries may coerce an agentic AI system into performing computationally expensive tool calls that waste resources and consume API budgets…
Bidda node: https://bidda.com/intelligence/mitre-atlas-agentic-resource-consumption
- AML.T0046 Spamming AI System with Chaff Data
This node addresses MITRE ATLAS technique AML.T0046 (Spamming AI System with Chaff Data). Adversaries may spam the AI system with chaff data that causes increase in the number of detections…
Bidda node: https://bidda.com/intelligence/mitre-atlas-spamming-ai-system-with-chaff-data
- AML.T0048 External Harms
This node addresses MITRE ATLAS technique AML.T0048.002 on AI-generated influence operations, aligning with EU AI Act Articles 9…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-influence-operations
- AML.T0048.000 Financial Harm
This node addresses MITRE ATLAS technique AML.T0048.000 (Financial Harm). Financial harm involves the loss of wealth, property, or other monetary assets due to theft, fraud or forgery, or pressure to provide financial resources to the…
Bidda node: https://bidda.com/intelligence/mitre-atlas-financial-harm
- AML.T0048.001 Reputational Harm
This node addresses MITRE ATLAS technique AML.T0048.001 (Reputational Harm). Reputational harm involves a degradation of public perception and trust in organizations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-reputational-harm
- AML.T0048.002 Societal Harm
This node addresses MITRE ATLAS technique AML.T0048.002 (Societal Harm). Societal harms might generate harmful outcomes that reach either the general public or specific vulnerable groups such as the exposure of children to vulgar content…
Bidda node: https://bidda.com/intelligence/mitre-atlas-societal-harm
- AML.T0048.003 User Harm
This node addresses MITRE ATLAS technique AML.T0048.003 (User Harm). User harms may encompass a variety of harm types including financial and reputational that are directed at or felt by individual victims of the attack rather than at the…
Bidda node: https://bidda.com/intelligence/mitre-atlas-user-harm
- AML.T0048.004 AI Intellectual Property Theft
This node addresses MITRE ATLAS technique AML.T0048.004 (AI Intellectual Property Theft). Adversaries may exfiltrate AI artifacts to steal intellectual property and cause economic harm to the victim organization…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-intellectual-property-theft
- AML.T0059 Erode Dataset Integrity
This node addresses adversarial techniques under MITRE ATLAS AML.T0059 (Erode Dataset Integrity), focusing on dataset poisoning and…
Bidda node: https://bidda.com/intelligence/mitre-atlas-erode-dataset-integrity
- AML.T0101 Data Destruction via AI Agent Tool Invocation
This node addresses MITRE ATLAS technique AML.T0107 on Defense Evasion via AI Agent Actions, aligning with EU AI Act Articles 9 and 15…
Bidda node: https://bidda.com/intelligence/mitre-atlas-defense-evasion-agent
- AML.T0112 Machine Compromise
This node addresses MITRE ATLAS technique AML.T0112 (Machine Compromise). Adversaries may compromise a machine by exploiting or manipulating AI-enabled components on the system…
Bidda node: https://bidda.com/intelligence/mitre-atlas-machine-compromise
- AML.T0112.000 Local AI Agent
This node addresses MITRE ATLAS technique AML.T0112.000 (Local AI Agent). Adversaries may achieve full system compromise by abusing AI agents running locally on a host, such as computer-use agents or AI-driven browsers…
Bidda node: https://bidda.com/intelligence/mitre-atlas-local-ai-agent
- AML.T0112.001 AI Artifacts
This node addresses MITRE ATLAS technique AML.T0112.001 (AI Artifacts). Adversaries may achieve full system compromise by introducing malicious AI artifacts, such as models or data, that contain embedded malware or other malicious…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-artifacts
[AML.TA0012] Privilege Escalation
- AML.T0012 Valid Accounts
This node addresses compliance obligations to mitigate MITRE ATLAS technique AML.T0012 (Valid ML Platform Account Abuse) for unauthorised AI…
Bidda node: https://bidda.com/intelligence/mitre-atlas-valid-ml-platform-accounts
- AML.T0053 AI Agent Tool Invocation
This node addresses MITRE ATLAS technique AML.T0053 (AI Agent Tool Invocation). Adversaries may use their access to an AI agent to invoke tools the agent has access to…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-agent-tool-invocation
- AML.T0054 LLM Jailbreak
This node addresses MITRE ATLAS technique AML.T0054 (LLM Jailbreak), focusing on adversarial prompt engineering to bypass AI safety controls, with…
Bidda node: https://bidda.com/intelligence/mitre-atlas-llm-jailbreak
- AML.T0105 Escape to Host
This node addresses MITRE ATLAS technique AML.T0081, focusing on persistence via compromised AI agents, under the EU AI Act…
Bidda node: https://bidda.com/intelligence/mitre-atlas-persistence-via-agent
[AML.TA0013] Credential Access
- AML.T0055 Unsecured Credentials
This node addresses MITRE ATLAS technique AML.T0055 (Unsecured Credentials). Adversaries may search compromised systems to find and obtain insecurely stored credentials…
Bidda node: https://bidda.com/intelligence/mitre-atlas-unsecured-credentials
- AML.T0082 RAG Credential Harvesting
This node addresses MITRE ATLAS technique AML.T0082 (RAG Credential Harvesting). Adversaries may attempt to use their access to a large language model (LLM) on the victim's system to collect credentials…
Bidda node: https://bidda.com/intelligence/mitre-atlas-rag-credential-harvesting
- AML.T0083 Credentials from AI Agent Configuration
This node addresses MITRE ATLAS technique AML.T0083 (Credentials from AI Agent Configuration). Adversaries may access the credentials of other tools or services on a system from the configuration of an AI agent…
Bidda node: https://bidda.com/intelligence/mitre-atlas-credentials-from-ai-agent-configuration
- AML.T0090 OS Credential Dumping
This node addresses MITRE ATLAS technique AML.T0090 (OS Credential Dumping). Adversaries may extract credentials from OS caches, application memory, or other sources on a compromised system…
Bidda node: https://bidda.com/intelligence/mitre-atlas-os-credential-dumping
- AML.T0098 AI Agent Tool Credential Harvesting
This node addresses adversarial techniques under MITRE ATLAS AML.T0098 (Credential Access via AI Agent Tools), focusing on harvesting credentials through autonomous AI…
Bidda node: https://bidda.com/intelligence/mitre-atlas-credential-access-via-agent
- AML.T0106 Exploitation for Credential Access
This node addresses MITRE ATLAS tactic AML.TA0008 for system and environment discovery via AI agents, aligning with EU…
Bidda node: https://bidda.com/intelligence/mitre-atlas-discovery-via-agent
[AML.TA0014] Command and Control
- AML.T0072 Reverse Shell
This node addresses MITRE ATLAS technique AML.T0072 (Reverse Shell). Adversaries may utilize a reverse shell to communicate and control the victim system…
Bidda node: https://bidda.com/intelligence/mitre-atlas-reverse-shell
- AML.T0096 AI Service API
This node addresses MITRE ATLAS technique AML.T0096 (AI Service API). Adversaries may communicate using the API of an AI service on the victim's system…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-service-api
- AML.T0108 AI Agent
This node addresses MITRE ATLAS technique AML.T0108 (AI Agent). Adversaries may abuse AI agents present on the victim's system for command and control…
Bidda node: https://bidda.com/intelligence/mitre-atlas-ai-agent
[AML.TA0015] Lateral Movement
- AML.T0052 Phishing
This node addresses MITRE ATLAS technique AML.T0052, focusing on spearphishing attacks targeting ML researchers, with compliance obligations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-phishing-ml-researcher
- AML.T0052.000 Spearphishing via Social Engineering LLM
This node addresses MITRE ATLAS technique AML.T0052.000 (Spearphishing via Social Engineering LLM). Adversaries may turn LLMs into targeted social engineers. LLMs are capable of interacting with users via text conversations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-spearphishing-via-social-engineering-llm
- AML.T0052.001 Deepfake-Assisted Phishing
This node addresses MITRE ATLAS technique AML.T0052.001 (Deepfake-Assisted Phishing). Adversaries may use deepfakes (AI-generated synthetic images, audio, or video) in phishing campaigns to impersonate trusted individuals, executives, or…
Bidda node: https://bidda.com/intelligence/mitre-atlas-deepfake-assisted-phishing
- AML.T0091 Use Alternate Authentication Material
This node addresses MITRE ATLAS technique AML.T0091 (Use Alternate Authentication Material). Adversaries may use alternate authentication material, such as password hashes, Kerberos tickets, and application access tokens, in order to move…
Bidda node: https://bidda.com/intelligence/mitre-atlas-use-alternate-authentication-material
- AML.T0091.000 Application Access Token
This node addresses MITRE ATLAS technique AML.T0091.000 (Application Access Token). Adversaries may use stolen application access tokens to bypass the typical authentication process and access restricted accounts, information, or services…
Bidda node: https://bidda.com/intelligence/mitre-atlas-application-access-token
[atlas-mitigations] Mitigations
Defensive controls (AML.M####)
- AML.M0000 Limit Public Release of Information
This node operationalises MITRE ATLAS mitigation AML.M0000 (Limit Public Release of Information). Limit the public release of technical information about the AI stack used in an organization's products or services…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-limit-public-release-of-information
- AML.M0001 Limit Model Artifact Release
This node operationalises MITRE ATLAS mitigation AML.M0001 (Limit Model Artifact Release). Limit public release of technical project details including data, algorithms, model architectures, and model checkpoints that are used in…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-limit-model-artifact-release
- AML.M0002 Passive AI Output Obfuscation
This node operationalises MITRE ATLAS mitigation AML.M0002 (Passive AI Output Obfuscation). Decreasing the fidelity of model outputs provided to the end user can reduce an adversary's ability to extract information about…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-passive-ai-output-obfuscation
- AML.M0003 Model Hardening
This node operationalises MITRE ATLAS mitigation AML.M0003 (Model Hardening). Use techniques to make AI models robust to adversarial inputs such as adversarial training or network…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-model-hardening
- AML.M0004 Restrict Number of AI Model Queries
This node operationalises MITRE ATLAS mitigation AML.M0004 (Restrict Number of AI Model Queries). Limit the total number and rate of queries a user can perform…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-restrict-number-of-ai-model-queries
- AML.M0005 Control Access to AI Models and Data at Rest
This node operationalises MITRE ATLAS mitigation AML.M0005 (Control Access to AI Models and Data at Rest). Establish access controls on internal model registries and limit internal access to production models…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-control-access-to-ai-models-and-data-at-rest
- AML.M0006 Use Ensemble Methods
This node operationalises MITRE ATLAS mitigation AML.M0006 (Use Ensemble Methods). Use an ensemble of models for inference to increase robustness to adversarial inputs…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-use-ensemble-methods
- AML.M0007 Sanitize Training Data
This node operationalises MITRE ATLAS mitigation AML.M0007 (Sanitize Training Data). Detect and remove or remediate poisoned training data…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-sanitize-training-data
- AML.M0008 Validate AI Model
This node operationalises MITRE ATLAS mitigation AML.M0008 (Validate AI Model). Validate that AI models perform as intended by testing for backdoor triggers, potential for data leakage, or adversarial influence…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-validate-ai-model
- AML.M0009 Use Multi-Modal Sensors
This node operationalises MITRE ATLAS mitigation AML.M0009 (Use Multi-Modal Sensors). Incorporate multiple sensors to integrate varying perspectives and modalities to avoid a single point of failure…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-use-multi-modal-sensors
- AML.M0010 Input Restoration
This node operationalises MITRE ATLAS mitigation AML.M0010 (Input Restoration). Preprocess all inference data to nullify or reverse potential adversarial perturbations…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-input-restoration
- AML.M0011 Restrict Library Loading
This node operationalises MITRE ATLAS mitigation AML.M0011 (Restrict Library Loading). Prevent abuse of library loading mechanisms in the operating system and software to load untrusted code by configuring appropriate library loading…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-restrict-library-loading
- AML.M0012 Encrypt Sensitive Information
This node operationalises MITRE ATLAS mitigation AML.M0012 (Encrypt Sensitive Information). Encrypt sensitive data such as AI models to protect against adversaries attempting to access…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-encrypt-sensitive-information
- AML.M0013 Code Signing
This node operationalises MITRE ATLAS mitigation AML.M0013 (Code Signing). Enforce binary and application integrity with digital signature verification to prevent untrusted code from executing…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-code-signing
- AML.M0014 Verify AI Artifacts
This node operationalises MITRE ATLAS mitigation AML.M0014 (Verify AI Artifacts). Verify the cryptographic checksum of all AI artifacts to verify that the file was not modified by an…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-verify-ai-artifacts
- AML.M0015 Adversarial Input Detection
This node operationalises MITRE ATLAS mitigation AML.M0015 (Adversarial Input Detection). Detect and block adversarial inputs or atypical queries that deviate from known benign behavior, exhibit behavior patterns observed in previous…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-adversarial-input-detection
- AML.M0016 Vulnerability Scanning
This node operationalises MITRE ATLAS mitigation AML.M0016 (Vulnerability Scanning). Vulnerability scanning is used to find potentially exploitable software vulnerabilities to remediate them…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-vulnerability-scanning
- AML.M0017 AI Model Distribution Methods
This node operationalises MITRE ATLAS mitigation AML.M0017 (AI Model Distribution Methods). Deploying AI models to edge devices can increase the attack surface of the system…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-ai-model-distribution-methods
- AML.M0018 User Training
This node operationalises MITRE ATLAS mitigation AML.M0018 (User Training). Educate AI model developers to on AI supply chain risks and potentially malicious AI artifacts…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-user-training
- AML.M0019 Control Access to AI Models and Data in Production
This node operationalises MITRE ATLAS mitigation AML.M0019 (Control Access to AI Models and Data in Production). Require users to verify their identities before accessing a production model…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-control-access-to-ai-models-and-data-in-production
- AML.M0020 Generative AI Guardrails
This node operationalises MITRE ATLAS mitigation AML.M0020 (Generative AI Guardrails). Guardrails are safety controls that are placed between a generative AI model and the output shared with the user to prevent undesired inputs and…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-generative-ai-guardrails
- AML.M0021 Generative AI Guidelines
This node operationalises MITRE ATLAS mitigation AML.M0021 (Generative AI Guidelines). Guidelines are safety controls that are placed between user-provided input and a generative AI model to help direct the model to produce desired…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-generative-ai-guidelines
- AML.M0022 Generative AI Model Alignment
This node operationalises MITRE ATLAS mitigation AML.M0022 (Generative AI Model Alignment). When training or fine-tuning a generative AI model it is important to utilize techniques that improve model alignment with safety, security, and…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-generative-ai-model-alignment
- AML.M0023 AI Bill of Materials
This node operationalises MITRE ATLAS mitigation AML.M0023 (AI Bill of Materials). An AI Bill of Materials (AI BOM) contains a full listing of artifacts and resources that were used in building the AI…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-ai-bill-of-materials
- AML.M0024 AI Telemetry Logging
This node operationalises MITRE ATLAS mitigation AML.M0024 (AI Telemetry Logging). Implement logging of inputs and outputs of deployed AI models…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-ai-telemetry-logging
- AML.M0025 Maintain AI Dataset Provenance
This node operationalises MITRE ATLAS mitigation AML.M0025 (Maintain AI Dataset Provenance). Maintain a detailed history of datasets used for AI applications…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-maintain-ai-dataset-provenance
- AML.M0026 Privileged AI Agent Permissions Configuration
This node operationalises MITRE ATLAS mitigation AML.M0026 (Privileged AI Agent Permissions Configuration). AI agents may be granted elevated privileges above that of a normal user to enable desired workflows…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-privileged-ai-agent-permissions-configuration
- AML.M0027 Single-User AI Agent Permissions Configuration
This node operationalises MITRE ATLAS mitigation AML.M0027 (Single-User AI Agent Permissions Configuration). When deploying an AI agent that acts as a representative of a user and performs actions on their behalf, it is important to…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-single-user-ai-agent-permissions-configuration
- AML.M0028 AI Agent Tools Permissions Configuration
This node operationalises MITRE ATLAS mitigation AML.M0028 (AI Agent Tools Permissions Configuration). When deploying tools that will be shared across multiple AI agents, it is important to implement robust policies and controls on…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-ai-agent-tools-permissions-configuration
- AML.M0029 Human In-the-Loop for AI Agent Actions
This node operationalises MITRE ATLAS mitigation AML.M0029 (Human In-the-Loop for AI Agent Actions). Systems should require the user or another human stakeholder to approve AI agent actions before the agent takes them…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-human-in-the-loop-for-ai-agent-actions
- AML.M0030 Restrict AI Agent Tool Invocation on Untrusted Data
This node operationalises MITRE ATLAS mitigation AML.M0030 (Restrict AI Agent Tool Invocation on Untrusted Data). Untrusted data can contain prompt injections that invoke an AI agent's tools, potentially causing confidentiality, integrity…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-restrict-ai-agent-tool-invocation-on-untrusted-data
- AML.M0031 Memory Hardening
This node operationalises MITRE ATLAS mitigation AML.M0031 (Memory Hardening). Memory Hardening involves developing trust boundaries and secure processes for how an AI agent stores and accesses memory and context…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-memory-hardening
- AML.M0032 Segmentation of AI Agent Components
This node operationalises MITRE ATLAS mitigation AML.M0032 (Segmentation of AI Agent Components). Define security boundaries around agentic tools and data sources with methods such as API access, container isolation, code execution…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-segmentation-of-ai-agent-components
- AML.M0033 Input and Output Validation for AI Agent Components
This node operationalises MITRE ATLAS mitigation AML.M0033 (Input and Output Validation for AI Agent Components). Implement validation on inputs and outputs for the tools and data sources used by AI agents…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-input-and-output-validation-for-ai-agent-components
- AML.M0034 Deepfake Detection
This node operationalises MITRE ATLAS mitigation AML.M0034 (Deepfake Detection). Apply deepfake detection algorithms against any untrusted or user-provided data, especially in impactful applications such as biometric verification, to…
Bidda node: https://bidda.com/intelligence/mitre-atlas-mitigation-deepfake-detection
Full interactive matrix: https://bidda.com/atlas
MITRE source: https://atlas.mitre.org/