AI Safety

AI Alignment

The research field focused on ensuring AI systems act in accordance with human values, intentions, and ethical principles.

AI Governance

The frameworks, policies, and regulations that guide the responsible development and deployment of AI technologies.

AI Explainability

The ability to describe AI decision-making processes in human-understandable terms, enabling trust and regulatory compliance.

AI Interpretability

The degree to which humans can understand the internal mechanics and reasoning of an AI model's predictions and decisions.

AI Fairness

The principle of ensuring AI systems make equitable decisions without discriminating against any group based on protected attributes.

AI Transparency

The practice of making AI systems' operations, data usage, and decision processes openly visible to stakeholders.

AI Robustness

The ability of an AI system to maintain performance under varying conditions, adversarial attacks, or noisy input data.

AI Hallucination

When an AI model generates plausible-sounding but factually incorrect or fabricated information with high confidence.

AI Red Teaming

The systematic adversarial testing of AI systems to identify vulnerabilities, failure modes, harmful outputs, and safety risks before deployment.

AI Watermarking

Techniques for embedding imperceptible statistical patterns in AI-generated content to enable reliable detection and provenance tracking of synthetic outputs.

AI Guardrails

Safety mechanisms and constraints implemented around AI systems to prevent harmful, biased, or policy-violating outputs while preserving useful functionality.

AI Model Card

A documentation framework that provides standardised information about an AI model's intended use, performance characteristics, limitations, and ethical considerations.

The economic model governing the pricing and allocation of computational resources for AI inference, including per-token billing, rate limiting, and credit systems.

An inference strategy that starts with known facts and applies rules to derive new conclusions until a goal is reached.

The ratio of true positive predictions to all positive predictions, measuring accuracy of positive classifications.

Area Under the ROC Curve, a single metric summarising a classifier's ability to distinguish between classes.

Techniques for combining the weights and capabilities of multiple fine-tuned models into a single model without additional training, creating versatile multi-capability systems.

A technique where a smaller, simpler model is trained to replicate the behaviour of a larger, more complex model.

A specialised processor designed to accelerate neural network computations in edge devices and mobile platforms.

A semiconductor designed specifically for AI and machine learning computations, optimised for parallel processing and matrix operations.