Today's auction is open. Place a bid →

AI & Crypto Glossary

A comprehensive glossary of terms covering artificial intelligence, cryptocurrency, and their intersection. 39 terms defined and explained.

AI & Machine Learning

AI Agent
An autonomous software program powered by artificial intelligence that can perceive its environment, make decisions, and take actions to achieve specific goals without continuous human intervention. AI agents can browse the web, execute code, manage files, and interact with APIs.
Large Language Model (LLM)
A type of artificial intelligence model trained on vast amounts of text data that can understand and generate human-like text. Examples include GPT-4, Claude, Llama, and Gemini. LLMs form the backbone of most modern AI assistants and chatbots.
Transformer
A neural network architecture introduced in the 2017 paper "Attention Is All You Need" that uses self-attention mechanisms to process sequential data. Transformers are the foundation of modern LLMs and have revolutionized natural language processing.
Fine-Tuning
The process of further training a pre-trained AI model on a specific dataset to specialize it for particular tasks or domains. Fine-tuning allows adapting general-purpose models to specific use cases with relatively little data.
RAG (Retrieval-Augmented Generation)
A technique that enhances LLM responses by first retrieving relevant documents from a knowledge base, then using them as context for generation. RAG reduces hallucinations and allows models to access up-to-date information.
Prompt Engineering
The practice of crafting effective instructions (prompts) for AI models to produce desired outputs. Good prompt engineering involves clear instructions, examples, context, and structured formatting to guide model behavior.
Hallucination
When an AI model generates information that sounds plausible but is factually incorrect or entirely fabricated. Hallucinations are a known limitation of LLMs and can be mitigated through techniques like RAG and grounding.
Context Window
The maximum amount of text (measured in tokens) that an AI model can process in a single interaction. Larger context windows allow models to handle longer documents and maintain more conversation history. Modern models range from 4K to 200K+ tokens.
Token
The basic unit of text processed by language models. A token can be a word, part of a word, or a punctuation mark. On average, one token equals about 4 characters or 0.75 words in English.
Embedding
A numerical vector representation of text, images, or other data that captures semantic meaning. Embeddings enable similarity search, clustering, and are fundamental to RAG systems and recommendation engines.
Multi-Modal AI
AI systems that can process and generate multiple types of data, such as text, images, audio, and video. Examples include GPT-4V (text + images) and Gemini (text + images + audio + video).
AI Crawler
A web bot operated by AI companies to scrape and index web content for training data or real-time retrieval. Examples include GPTBot (OpenAI), ClaudeBot (Anthropic), and PerplexityBot. Site owners can control access via robots.txt.
Inference
The process of running a trained AI model on new input data to generate predictions or outputs. Inference is the production-use phase of a model, as opposed to training. Inference speed and cost are key factors in deploying AI at scale.
RLHF (Reinforcement Learning from Human Feedback)
A training technique where an AI model is refined using human preference data. Annotators rank model outputs, and the model learns to produce responses that humans prefer. RLHF is central to aligning LLMs like ChatGPT and Claude with human values.
Model Distillation
A technique where a smaller "student" model is trained to replicate the behavior of a larger "teacher" model. Distillation produces compact models that retain much of the capability of the original while being faster and cheaper to run.
Zero-Shot Learning
The ability of an AI model to perform a task it was not explicitly trained for, by leveraging its general knowledge. For example, a model can classify text into categories it has never seen by understanding the category descriptions.
Agentic AI
AI systems designed to act autonomously over extended periods, making decisions, using tools, and completing multi-step workflows without continuous human guidance. Agentic AI goes beyond question-answering to pursue goals independently.
MCP (Model Context Protocol)
An open protocol that standardizes how AI models connect to external tools and data sources. MCP provides a universal interface for LLMs to access databases, APIs, file systems, and other services through a consistent client-server architecture.

Cryptocurrency & Blockchain

Blockchain
A decentralized, distributed digital ledger that records transactions across many computers in a way that makes the records resistant to modification. Each block contains a cryptographic hash of the previous block, creating a chain.
Smart Contract
Self-executing programs stored on a blockchain that automatically enforce the terms of an agreement when predetermined conditions are met. Smart contracts enable trustless transactions and form the basis of DeFi and DAOs.
Cryptocurrency Wallet
Software or hardware that stores private keys and allows users to send, receive, and manage cryptocurrency. Wallets can be hot (connected to internet) or cold (offline). Examples include MetaMask, Phantom, and Ledger.
DeFi (Decentralized Finance)
Financial services built on blockchain technology that operate without traditional intermediaries like banks. DeFi protocols enable lending, borrowing, trading, and yield farming through smart contracts.
DAO (Decentralized Autonomous Organization)
An organization represented by rules encoded as smart contracts on a blockchain. Members collectively make decisions through token-based voting, without centralized leadership or management hierarchy.
Gas Fee
The cost required to execute transactions or smart contracts on a blockchain network like Ethereum. Gas fees compensate miners/validators for the computational resources needed to process and validate transactions.
Web3
The concept of a decentralized internet built on blockchain technology, where users own their data and digital assets. Web3 encompasses cryptocurrencies, NFTs, DeFi, DAOs, and decentralized applications (dApps).
Stablecoin
A cryptocurrency designed to maintain a stable value by pegging it to a reference asset like the US dollar. Examples include USDC, USDT, and DAI. Stablecoins are widely used in DeFi and cross-border payments.
Layer 2 (L2)
A secondary framework or protocol built on top of an existing blockchain (Layer 1) to improve scalability and reduce transaction costs. Examples include Optimism, Arbitrum, and Lightning Network. L2s are critical for enabling AI micro-payments on-chain.
Oracle
A service that provides external real-world data to smart contracts on a blockchain. Oracles bridge the gap between on-chain and off-chain environments, enabling smart contracts to react to real-world events like price changes or AI model outputs.
Proof of Work vs Proof of Stake
Two consensus mechanisms for validating blockchain transactions. Proof of Work (Bitcoin) requires miners to solve computational puzzles. Proof of Stake (Ethereum post-merge) requires validators to lock up tokens as collateral. PoS uses far less energy.

AI x Crypto

AI x Crypto
The emerging intersection of artificial intelligence and blockchain technology. Use cases include decentralized AI model training, AI-powered trading bots, on-chain AI agents, token-gated AI services, and AI-verified blockchain transactions.
On-Chain AI Agent
An AI agent that can interact with blockchain networks autonomously—executing transactions, managing wallets, participating in governance votes, and interacting with smart contracts without human intervention.
AI Model Marketplace
A platform where AI models, training data, or inference compute can be bought, sold, or rented using cryptocurrency. These marketplaces aim to democratize access to AI while compensating creators fairly.
Decentralized AI
AI systems that distribute training, inference, or governance across multiple nodes rather than relying on centralized servers. Projects like Bittensor and Akash aim to create decentralized networks for AI computation.
Tokenized AI Services
AI capabilities offered as blockchain-based services where access is controlled through token ownership or micro-payments in cryptocurrency. This model enables pay-per-use AI without traditional subscriptions.
AI-Powered Trading Bot
An autonomous program that uses machine learning models to analyze market data and execute cryptocurrency trades. These bots can process vast amounts of data, identify patterns, and trade 24/7 without human intervention, operating at speeds impossible for manual traders.
agents.txt
A proposed standard file (similar to robots.txt) that declares a website's capabilities for autonomous AI agents. Unlike robots.txt which controls access, agents.txt advertises what agents can do on the site — available APIs, payment methods, and interaction protocols.
USDC Auction
A time-bounded bidding mechanism where participants compete using USDC (a dollar-pegged stablecoin) for limited resources like ad placements. Auctions reset daily and the highest bidder wins. Both humans and AI agents can participate.
Agent Registry
A directory of known autonomous AI agents with their capabilities, blockchain affiliations, and wallet addresses. Enables agent discovery and reputation tracking across the ecosystem.
Bid Floor
The minimum bid amount accepted in an auction. In AI agent ad auctions, the bid floor prevents spam bids and establishes a baseline value for advertising inventory.