11 new AI terms you need to understand

Prompt Engineering

The process of crafting inputs or prompts to get the best possible response from an AI system.

Asking ChatGPT to write an email, by giving it specific instructions like tone and subject.

Large Language Model

A type of AI trained on vast amounts of text to understand and generate human-like language.

ChatGPT is a large language model capable of answering questions and holding conversations.

Natural Language Processing (NLP)

The technology that enables computers to understand and work with human language.

Using voice assistants like Siri or Alexa to answer everyday questions.

Machine Learning

A method where computers learn patterns from data to make decisions or predictions without being explicitly programmed.

Netflix recommending shows you may like, based on your viewing history.

Neural Network

A type of machine learning inspired by the human brain, consisting of layers of nodes that process data.

Facial recognition systems that identify people from photos.

Generative AI

AI that creates new content, such as text, images, or music, based on the data it was trained on.

AI tools like DALL-E generating artwork or images from text descriptions.

Reinforcement Learning

A method where AI learns by trial and error, getting rewards or penalties to improve its decisions.

AI in self-driving cars learning how to navigate roads by receiving feedback from its performance.

Supervised Learning

Training an AI model using labeled data, where the correct output is provided for each input during training.

Teaching an AI to recognize cats by showing it labeled images of cats and non-cats.

Unsupervised Learning

Training an AI model with unlabeled data, where the AI looks for patterns or groupings on its own.

AI clustering customer behavior data to identify new segments without knowing what they are in advance.

Deep Learning

A subset of machine learning using neural networks with many layers to process large and complex data.

AI that powers speech recognition in smartphones or automatic translation tools.

Hallucinations

Refer to instances where an AI system generates incorrect, misleading, or entirely fabricated information that is not based on the data it was trained on or the input it received. This often happens with language models like ChatGPT, which may confidently provide false facts, create non-existent references, or invent responses that sound plausible but are not accurate.

These errors occur because the AI generates content by predicting the next word in a sequence based on patterns from its training data, but it doesn’t actually ‘know’ facts in the way humans do.

Input: "Who won the 2023 Nobel Prize in Physics?"

AI Hallucination: "The 2023 Nobel Prize in Physics was awarded to Dr. Elena Johnson for her ground-breaking work on dark matter teleportation."

In this case, the AI might invent a person, a discovery, or an award that never actually happened, confidently presenting the fabricated information as if it were true. This happens because the AI generates responses by predicting patterns rather than verifying facts.