Understand the ideas behind the news.
Plain-English explainers for the concepts and techniques shaping AI — each one tied to the latest news as it happens.
60 concepts explained
Start here
AI Agents
LLM systems that plan, call tools, and act over multiple steps toward a goal.
Read the conceptConcepts
Model Context Protocol
An open standard for connecting AI assistants to external tools and data sources.
Concept· 59 stories
Mixture of Experts
A model design that routes each input to a subset of specialised sub-networks for efficiency.
Concept· 15 stories
Transformers
The neural-network architecture, built on attention, behind nearly all modern LLMs.
Concept· 11 stories
Diffusion Models
Generative models that create images or audio by iteratively denoising random noise.
Concept· 6 stories
Embeddings
Numeric vector representations of text or data that let machines compare meaning by distance.
Concept· 1 story
Inference
Running a trained model to produce outputs — the compute cost users actually pay at runtime.
Concept· 1 story
Vector Database
A database built to store and search embeddings by similarity, the backbone of RAG systems.
Concept
Context Window
The maximum amount of text a model can consider at once when generating a response.
Concept
India
Concept· 115 stories
Python
Concept· 39 stories
Agentic AI
Concept· 22 stories
Rust
Concept· 20 stories
macOS
Concept· 18 stories
Sam Altman
Concept· 15 stories
robotics
Concept· 13 stories
Reinforcement Learning
Concept· 13 stories
JavaScript
Concept· 12 stories
semiconductors
Concept· 11 stories
GPU
Concept· 11 stories
open-source
Concept· 10 stories
China
Concept· 9 stories
autonomous agents
Concept· 8 stories
AI
Concept· 8 stories
WebAssembly
Concept· 7 stories
Jalapeño
Concept· 7 stories
Physical AI
Concept· 7 stories
multi-agent systems
Concept· 7 stories
embodied AI
Concept· 7 stories
Elon Musk
Concept· 6 stories
iOS
Concept· 6 stories
AI Governance
Concept· 6 stories
sovereign AI
Concept· 6 stories
AI infrastructure
Concept· 6 stories
AI coding agents
Concept· 5 stories
Apple Silicon
Concept· 5 stories
Dario Amodei
Concept· 5 stories
AGI
Concept· 5 stories
world models
Concept· 5 stories
open-weight models
Concept· 5 stories
Linux
Concept· 5 stories
computer vision
Concept· 5 stories
coding-agents
Concept· 5 stories
LLM agents
Concept· 5 stories
Gemini API
Concept· 5 stories
AI safety
Concept· 4 stories
Apple Intelligence
Concept· 4 stories
data centers
Concept· 4 stories
Pax Silica
Concept· 4 stories
Digital Public Infrastructure
Concept· 4 stories
Edge AI
Concept· 4 stories
drug discovery
Concept· 4 stories
Google I/O
Concept· 4 stories
Techniques
Retrieval-Augmented Generation
A technique that grounds an LLM's answers in documents retrieved at query time, reducing hallucination.
Technique· 17 stories
Fine-Tuning
Further training a pretrained model on a narrower dataset to specialise its behaviour.
Technique· 5 stories
Prompt Engineering
The practice of crafting model inputs to reliably get the outputs you want.
Technique· 3 stories
Chain of Thought
Prompting a model to reason step by step, which improves accuracy on complex tasks.
Technique· 1 story
RLHF
Reinforcement Learning from Human Feedback — tuning models to match human preferences.
Technique· 1 story
Quantization
Compressing a model's weights to lower precision so it runs faster and on less memory.
Technique
prompt injection
Technique· 9 stories