Generative AI
AI systems that generate new text, images, audio, code, or other content from learned patterns.
Generative AI
A category of artificial intelligence systems capable of producing new content such as text, images, code, or audio based on patterns learned from large datasets.
What Is Generative AI
Generative AI refers to models designed to create novel outputs rather than only classify or retrieve existing data. Instead of simply identifying patterns, these systems can generate coherent language, synthesize visuals, draft code, and produce audio that resembles human-created material.
How Generative AI Models Work
Generative models are trained on large corpora to learn statistical relationships between tokens, pixels, or sound segments. During inference, they predict the most likely next unit and iteratively build full outputs. Architectures such as transformers and diffusion models enable these systems to model complex context and produce fluent, high-detail generations.
Types of Generative AI Systems
- Large language models (LLMs): generate and transform text.
- Image generation models: create images from prompts or examples.
- Code generation models: assist with programming and documentation.
- Speech and audio models: generate voice, sound effects, or music.
- Multimodal models: operate across text, vision, and audio together.
Applications in Language Technologies and Translation
In translation workflows, generative AI supports drafting, rewriting, summarisation, terminology adaptation, and multilingual content creation. It can use document-level context to improve consistency and tone, while still requiring human validation for accuracy, domain specificity, and regulatory compliance.
Benefits and Challenges of Generative AI
Benefits include speed, scalability, and support for creative or repetitive language tasks. Key challenges include hallucinations, bias, data privacy concerns, and quality variability in specialised domains. Effective professional use depends on governance, glossary controls, and expert post-editing.
Related Terms
GDPR (General Data Protection Regulation)
EU legislation regulating the processing and protection of personal data.
Gender Bias in AI
Differences in how AI treats or represents genders due to training-data patterns.
GPU (Graphics Processing Unit)
A specialised processor that accelerates parallel computations for training and running AI models.
Gradient Descent
An optimisation algorithm that iteratively updates model parameters to reduce prediction error.
Glossary-driven Translation
Translation guided by predefined terminology lists.
Related Resources
Machine Learning
A field of AI where systems learn patterns from data to make predictions or generate content.
Multimodal AI Models
AI systems that combine text, image, audio, and video inputs to improve understanding and generation across tasks.
Neural Network
A computational model made up of interconnected layers that learns patterns from data and powers many modern AI systems.
OCR (Optical Character Recognition)
Technology that extracts text from scanned documents or images.
Artificial Intelligence
Computational systems capable of performing tasks that traditionally require human intelligence.
File Parsing
Automated extraction of text and structure from formats such as DOCX, PDF, PPTX, or XLSX.