AGI: Al that can think like humans.
Al Agents: Autonomous programs that make decisions.
Al Alignment: Ensuring Al follows human values.
Al Model: A trained system for a task.
Al Wrapper: Simplifies interaction with Al models.
Fine-tuning: Improving Al with specific training data.
Hallucination: When Al generates false information.
Chatbot: Al that simulates human conversation.
Compute: Processing power for Al models.
Computer Vision: Al that understands images and videos.
Context: Information Al retains for better responses.
CoT (Chain of Thought): Al thinking step-by-step.
Deep Learning: Al learning through layered neural networks.
Embedding: Numeric representation of words for Al.
Explainability: How Al decisions are understood.
Foundation Model: Large Al model adaptable to tasks.
Generative Al: Al that creates text, images, etc.
GPU: Hardware for fast Al processing.
Ground Truth: Verified data Al learns from.
Inference: Al making predictions on new data.
LLM (Large Language Model): Al trained on vast text data.
Machine Learning: Al improving from data experience.
MCP (Model Context Protocol): Standard for Al external data access.
NLP (Natural Language Processing): Al understanding human language.
Neural Network: Al model inspired by the brain.
Parameters: Al's internal variables for learning.
Prompt Engineering: Crafting inputs to guide Al output.
Reasoning Model: Al that follows logical thinking.
Reinforcement Learning: Al learning from rewards and penalties.
RAG (Retrieval-Augmented Generation): Al combining search with responses.
Supervised Learning: Al trained on labeled data.
TPU: Google's Al-specialized processor.
Tokenization: Breaking text into smaller parts.
Training: Teaching Al by adjusting its parameters.
Transformer: Al architecture for language processing.
Unsupervised Learning: Al finding patterns in unlabeled data.
Vibe Coding: Al-assisted coding via natural language prompts.
Weights: Values that shape Al learning.