Natural Language Processing (NLP)
AI techniques for understanding, interpreting, and generating human language. NLP encompasses sentiment analysis, named entity recognition, machine translation, text summarization, and conversational AI. It forms the foundation of modern language models and chatbot systems.
Why It Matters
NLP powers the AI interfaces most people interact with daily — search engines, virtual assistants, chatbots, translation tools. Governance considerations include bias in language understanding across dialects and demographics, privacy in text analysis, and accuracy in high-stakes text classification.
Example
An NLP-based content moderation system must handle nuance across languages, cultures, and contexts — sarcasm, coded language, context-dependent slang — or risk both over-censorship (silencing legitimate speech) and under-censorship (missing harmful content).
Think of it like...
NLP is like teaching a computer to be a polyglot translator who also understands tone, context, and subtext — it's not just about the words, but what they mean in this specific conversation.
Related Terms
Large Language Model (LLM)
A type of foundation model trained on massive text datasets that can understand, generate, and reason about human language. LLMs like GPT-4, Claude, and Gemini use transformer architecture and typically have billions of parameters, enabling capabilities from summarization to coding to complex reasoning.
Sentiment Analysis
The NLP task of identifying and classifying the emotional tone or opinion expressed in text as positive, negative, or neutral. Advanced systems detect nuanced emotions like frustration, excitement, or sarcasm.
Bias (AI)
Systematic errors in AI system outputs that produce unfair or skewed outcomes. AI bias can originate from training data (historical, representation, measurement, sampling, or aggregation bias), from model design choices, or from the deployment context. Bias is not always obvious and can compound through the AI lifecycle.