Model Extraction
Definition
Model extraction involves querying a target machine learning model’s API to infer …
Security
Watermarking
Definition
As generative AI produces increasing amounts of synthetic media, watermarking serves as a …
Sovereign AI
Definition
Sovereign AI describes the capability of a country or organization to build, deploy, and …
Slopaganda
Definition
Slopaganda describes a strategic form of disinformation that relies on repetition, …
Robustness
Definition
In AI safety and ethics, robustness refers to a model’s resilience against …
Resisting AI
Definition
Resisting AI refers to methods used by individuals or entities to avoid being influenced, …
Rate Limiting
Definition
Rate limiting protects AI services and APIs from abuse, overload, and excessive resource …
Novelty detection
Definition
Novelty detection is a machine learning task focused on identifying data points that do …
Native-language identification
Definition
Native-language identification (NLI) is a subfield of natural language processing that …
Inauthentic text
Definition
Inauthentic text refers to written material produced by AI systems or humans with …
Differential Privacy
Definition
Differential privacy provides strong privacy guarantees by adding calibrated statistical …
Data Poisoning
Definition
This adversarial technique aims to compromise the integrity of machine learning models by …
Cybersecurity
Definition
Cybersecurity encompasses the technologies, processes, and practices designed to protect …
Content Provenance
Definition
Content provenance refers to the documentation and verification of where digital content …
Anomaly detection
Definition
Anomaly detection, also known as outlier detection, involves analyzing data to find …
Adversarial Attack
Definition
Adversarial attacks exploit the vulnerabilities of neural networks by introducing subtle …
Adversarial machine learning
Definition
This field encompasses both offensive techniques to break models and defensive strategies …
AI Security Institute
Definition
An AI Security Institute is a specialized entity focused on mitigating risks associated …
Prompt Injection
Definition
Prompt injection exploits the way large language models interpret user instructions by …
Jailbreak
Definition
Jailbreaking involves crafting specific inputs or prompts that trick an AI model into …
Federated Learning
Definition
Federated learning enables organizations to collaboratively train AI models without …
Data Protection
Definition
Data Protection encompasses legal, technical, and organizational measures designed to …
Deepfake
Definition
Deepfakes are hyper-realistic audio or video manipulations created using generative …
Safe
Definition
Safety in AI involves implementing constraints and safeguards to ensure that automated …