Artificial intelligence is a deep and convoluted world. Scientists in this field often rely on jargon and lingo to explain what they are working on. As a result, we frequently have to use those technical terms in our coverage of the artificial intelligence industry.
That is why TechCrunch has put together a glossary with definitions of some of the most important words and phrases that appear in AI articles. Researchers continually uncover novel methods to push the frontier of artificial intelligence while identifying emerging safety risks. This guide will be regularly updated to keep pace with this fast-moving field.
Let’s break down the AI alphabet soup into plain English.
Read also: Google Chrome Installed a 4GB AI File on Your PC Without Asking. Here's What You Need to Know
Quick Facts Box
Glossary of Key AI Terms
AGI (Artificial General Intelligence)
AGI is a nebulous term. But it generally refers to AI that is more capable than the average human at many, if not most, tasks. OpenAI CEO Sam Altman has described AGI as the “equivalent of a median human that you could hire as a co‑worker.” Meanwhile, OpenAI’s charter defines AGI as “highly autonomous systems that outperform humans at most economically valuable work.” Google DeepMind views AGI as “AI that is at least as capable as humans at most cognitive tasks.” Confused? So are experts at the forefront of AI research.
AI Agent
An AI agent refers to a tool that uses AI technologies to perform a series of tasks on your behalf – beyond what a more basic AI chatbot could do. Examples include filing expenses, booking tickets or a table at a restaurant, or even writing and maintaining code. There are many moving pieces in this emergent space, so “AI agent” might mean different things to different people. But the basic concept implies an autonomous system that may draw on multiple AI systems to carry out multi‑step tasks.
Chain of Thought
Given a simple question, a human brain can answer without even thinking too much about it. But in many cases, you often need a pen and paper to arrive at the right answer because there are intermediate steps. In an AI context, chain‑of‑thought reasoning for large language models means breaking down a problem into smaller, intermediate steps to improve the quality of the end result. It usually takes longer to get an answer, but the answer is more likely to be correct, especially in logic or coding tasks.
Read also: Google’s AI drug discovery firm Isomorphic Labs is raising $2 billion to fight cancer.
Compute
Compute generally refers to the vital computational power that allows AI models to operate. This type of processing fuels the AI industry, giving it the ability to train and deploy its powerful models. The term is often a shorthand for the kinds of hardware that provide the computational power – GPUs, CPUs, TPUs, and other forms of infrastructure that form the bedrock of the modern AI industry.
Deep Learning
A subset of self‑improving machine learning in which AI algorithms are designed with a multi‑layered, artificial neural network structure. This allows them to make more complex correlations compared to simpler machine learning‑based systems. Deep learning AI models are able to identify important characteristics in data themselves, rather than requiring human engineers to define these features.
Diffusion
Diffusion is the tech at the heart of many art‑, music‑, and text‑generating AI models. Inspired by physics, diffusion systems slowly “destroy” the structure of data – photos, songs, and so on – by adding noise until there is nothing left. In physics, diffusion is spontaneous and irreversible. But diffusion systems in AI aim to learn a sort of “reverse diffusion” process to restore the destroyed data, gaining the ability to recover the data from noise.
Read also: The Instax Wide 400 Builds on Instant Photography’s Simplicity – and Stretches It, Literally
Distillation
A technique used to extract knowledge from a large AI model with a “teacher‑student” model. Developers send requests to a teacher model and record the outputs. These outputs are then used to train the student model, which learns to approximate the teacher’s behavior. Distillation can be used to create a smaller, more efficient model based on a larger model. While all AI companies use distillation internally, it may have also been used by some AI companies to catch up with frontier models.
Fine‑tuning
The further training of an AI model to optimize performance for a more specific task or area than was previously a focal point of its training – typically by feeding in new, specialized data. Many AI startups are taking large language models as a starting point to build commercial products but are vying to amp up utility for a target sector or task by supplementing earlier training cycles with fine‑tuning based on their own domain‑specific knowledge.
GAN (Generative Adversarial Network)
A machine learning framework that underpins some important developments in generative AI when it comes to producing realistic data – including deepfake tools. GANs involve a pair of neural networks: one generates an output, and the other evaluates it. This is set up as a competition, with the two models programmed to try to outdo each other. This structured contest can optimize AI outputs to be more realistic without the need for additional human intervention.
Hallucination
Hallucination is the AI industry’s preferred term for AI models making stuff up – literally generating incorrect information. Obviously, it is a huge problem for AI quality. Hallucinations produce outputs that can be misleading and could even lead to real‑life risks, with potentially dangerous consequences (think of a health query that returns harmful medical advice). This is why most AI tools’ small print now warns users to verify AI‑generated answers. The problem arises as a consequence of gaps in training data. Hallucinations are contributing to a push towards increasingly specialized and/or vertical AI models as a way to reduce the likelihood of knowledge gaps and shrink disinformation risks.
Inference
Inference is the process of running an AI model. It is setting a model loose to make predictions or draw conclusions from previously seen data. To be clear, inference cannot happen without training; a model must learn patterns in a set of data before it can effectively extrapolate from this training data. Many types of hardware can perform inference, ranging from smartphone processors to beefy GPUs to custom‑designed AI accelerators.
Large Language Model (LLM)
LLMs are the AI models used by popular AI assistants, such as ChatGPT, Claude, Google’s Gemini, Meta’s AI Llama, Microsoft Copilot, and Mistral’s Le Chat. When you chat with an AI assistant, you interact with an LLM that processes your request. LLMs are deep neural networks made of billions of numerical parameters that learn the relationships between words and phrases. These models are created from encoding the patterns they find in billions of books, articles, and transcripts. When you prompt an LLM, the model generates the most likely pattern that fits the prompt.
Memory Cache
Memory cache refers to an important process that boosts inference. In essence, caching is an optimization technique designed to make inference more efficient. Caching cuts down on the number of calculations a model might have to run by saving particular calculations for future user queries and operations. One well‑known example is KV (key value) caching, which works in transformer‑based models and drives faster results by reducing the time it takes to generate answers to user questions.
Neural Network
A neural network refers to the multi‑layered algorithmic structure that underpins deep learning – and more broadly, the whole boom in generative AI tools. Although the idea of taking inspiration from the interconnected pathways of the human brain as a design structure for data processing algorithms dates back to the 1940s, it was the rise of graphical processing hardware (GPUs) – via the video game industry – that truly unlocked the power of this theory. These chips enabled neural network‑based AI systems to achieve far better performance across many domains, including voice recognition, autonomous navigation, and drug discovery.
RAMageddon
RAMageddon is the fun new term for a not‑so‑fun trend sweeping the tech industry: an ever‑increasing shortage of random access memory chips, which power pretty much all the tech products we use daily. As the AI industry has blossomed, the biggest tech companies and AI labs – all vying to have the most powerful and efficient AI – are buying so much RAM to power their data centers that there is not much left for the rest of us. That supply bottleneck means what is left is getting more and more expensive, affecting industries like gaming and consumer electronics.
Read also: Apple Invests ₹100 Crore in India's Green Future: Why This Matters for Your Business
Training
Developing machine learning AIs involves a process known as training. In simple terms, this refers to data being fed in so the model can learn from patterns and generate useful outputs. It is only through training that the AI model really takes shape. It is important to note that not all AI requires training. Rules‑based AIs that are programmed to follow manually predefined instructions do not need to undergo training. However, such AI systems are likely to be more constrained than well‑trained self‑learning systems.
Tokens
Tokens represent the basic building blocks of human‑AI communication. They are discrete segments of data that have either been processed or produced by an LLM. Tokens are created via a process known as “tokenization,” which breaks down raw data and refines it into distinct units that are digestible to an LLM. There are several different kinds of tokens – input tokens, output tokens, and reasoning tokens. With enterprise AI, token usage determines costs. Most AI companies charge for LLM usage on a per‑token basis.
Transfer Learning
A technique where a previously trained AI model is used as the starting point for developing a new model for a different but typically related task – allowing knowledge gained in previous training cycles to be reapplied. Transfer learning can drive efficiency savings by shortcutting model development. It can also be useful when data for the task is somewhat limited. However, models that rely on transfer learning may require training on additional data to perform well in their domain of focus.
Read also: Crypto Exchange Coinbase Cuts 700 Jobs, Betting Everything on AI
Weights
Weights are core to AI training, as they determine how much importance (or weight) is given to different features in the data used for training the system – thereby shaping the AI model’s output. Weights are numerical parameters that define what is most salient in a dataset for the given training task. Model training typically begins with weights that are randomly assigned, but as the process unfolds, the weights adjust as the model seeks to arrive at an output that more closely matches the target.
FAQ
Q1: What is the difference between AI, machine learning, and deep learning?
A: Artificial intelligence is the broad field of creating intelligent machines. Machine learning is a subset of AI where systems learn from data without being explicitly programmed. Deep learning is a subset of machine learning that uses multi‑layered neural networks to process data in complex ways.
Q2: Why do AI models hallucinate?
A: Hallucinations occur because AI models have gaps in their training data. They are designed to generate patterns that fit a prompt, but when they lack the correct information, they may confidently produce plausible‑sounding but completely false answers. There is not enough data in existence to train models to comprehensively resolve all the questions we could possibly ask.
Q3: How are tokens different from words?
A: Tokens are not the same as words. A single word can be broken into multiple tokens, especially for longer or less common words. For example, the word “unbelievable” might be split into “un”, “believe”, and “able.” AI companies charge based on token count, not word count.
Q4: What is the difference between training and inference?
A: Training is the process of feeding data to an AI model so it can learn patterns and relationships. Inference is running a trained model to generate responses or predictions. Training is computationally expensive and happens once (or periodically). Inference happens every time you use an AI tool.
Q5: Is AGI the same as superintelligence?
A: No. AGI refers to AI that matches or exceeds human capabilities across most tasks. Superintelligence refers to AI that vastly surpasses the smartest human minds in virtually every field. Most researchers believe AGI would need to be achieved before superintelligence becomes possible.
Read also: Marc Lore Says AI Will Enable Anyone to Open a Restaurant in Under a Minute
Did this glossary help clarify AI terminology for you? Which term do you still find confusing? Drop your questions in the comments below – I will do my best to answer them.
If you found this guide useful, share it with a colleague who is just starting to explore artificial intelligence. Understanding the language is the first step to understanding the technology.
Tags: AI Glossary, Artificial Intelligence, LLM, Hallucination, AGI, Machine Learning, AI Terminology, Tech Guide
Read also: MCP vs API: What Every Developer Needs to Know About AI Agent Infrastructure in 2026
Have a question about AI or the latest tech trends? We’d love to hear your thoughts!
Please stay on topic and keep it helpful. Note: All comments are moderated to keep our community spam-free.