Legal frameworks, regulatory standards, and ethical principles governing the development, deployment, and oversight of artificial intelligence systems. This category covers the intersection of AI with compliance, intellectual property, and the justice system.
| Term | Description |
|---|---|
| AI Act | Comprehensive EU regulation establishing a risk-based framework for the development, deployment, and use of artificial intelligence systems. |
| AI Copyright & IP | Legal framework governing the ownership and permissible use of AI-generated outputs and the copyrighted data used to train models. |
| AI Forensics (Media Forensics) | Scientific process of analyzing digital data to detect AI-generated or manipulated content, such as deepfakes, for legal admissibility. |
| AI-Generated Evidence | Digital evidence created or materially altered by AI, presenting complex challenges for authentication and admissibility in court. |
| AI Liability | Legal principles determining who is responsible for damages or harms caused by the actions or decisions of an AI system. |
| Algorithmic Accountability | The obligation of organizations to take responsibility for the outcomes produced by their AI systems and address any negative impacts. |
| Algorithmic Audit | Systematic evaluation of an AI system’s code, data, and outcomes to ensure compliance with ethical, legal, and performance standards. |
| Algorithmic Risk Assessment | Use of predictive algorithms, primarily in criminal justice, to evaluate an individual’s likelihood of future offending or flight risk. |
| Automated Decision-Making (ADM) | Systems that use algorithms and data to make or significantly influence decisions affecting individuals without human intervention. |
| eDiscovery (Electronic Discovery) | Legally defensible process of identifying, collecting, and producing Electronically Stored Information (ESI) for litigation or investigations. |
| Generative AI Disclosure | Legal or regulatory requirement to transparently inform users when they are interacting with AI or consuming AI-generated content. |
| High-Risk AI System | AI applications posing significant threats to health, safety, or fundamental rights, subject to strict compliance requirements. |
| Impact Assessment (Algorithmic) | Evaluation process to identify and mitigate potential negative effects of an AI system on individuals’ rights and freedoms before deployment. |
| Regulatory Sandbox | Controlled environment created by regulators allowing businesses to test innovative AI products under relaxed regulatory supervision. |
| Right to Explanation | Legal principle granting individuals the right to receive meaningful information about the logic of automated decisions affecting them. |
| Systemic Risk | Potential for widespread failure or cascading negative impacts across financial, social, or technological systems caused by AI integration. |
| TAR (Technology-Assisted Review) | Legal process in eDiscovery using machine learning to prioritize, classify, and review large volumes of documents for relevance. |
| Trustworthy AI | Framework and design principles ensuring AI systems are lawful, ethical, and robust, prioritizing human oversight and societal well-being. |