AI basics
What is generative AI?
Generative AI creates new content such as text, images, audio and code from a prompt. Learn how it works, what it can make and where it falls short.
Generative AI is a type of artificial intelligence that creates new content, such as text, images, audio, video, or code, based on a prompt you give it. Instead of only sorting or predicting from existing data, it produces something that did not exist before, in the style and structure of what it learned from.
It sits inside the wider field of artificial intelligence and is built with machine learning.
How does generative AI work?
A generative model is trained on a large collection of examples, such as text or images. During training it learns the statistical patterns of that material: which words tend to follow which, or what shapes and textures make up a "dog on a beach".
When you give it a prompt, it uses those learned patterns to produce a new output that fits. Two common approaches are:
- Language models, which generate text one token at a time. See what is a large language model.
- Image models, such as diffusion models, which start from random noise and gradually refine it into a picture that matches the prompt. See what is AI image generation.
What can generative AI create?
| Output | Examples |
|---|---|
| Text | Answers, emails, summaries, stories |
| Images | Illustrations, concept art, edits to a photo |
| Audio | Synthetic speech, music |
| Code | Functions, scripts, explanations |
| Video | Short generated clips |
Generative AI vs traditional AI
Traditional or predictive AI answers a question about existing data: is this email spam, what will sales be next month. Generative AI produces new material. The two use similar underlying machine learning, but the output differs: a label or number versus a piece of content.
What are the risks and limits?
- Inaccuracy. Generated text can contain errors presented confidently. See what are AI hallucinations.
- Bias. Outputs reflect patterns, including skewed ones, in training data.
- Originality and rights. Questions about copyright and training data are still being argued in courts and policy debates, so check rules before commercial use.
- Misuse. The same tools can make misleading content, so disclose AI use where it matters.
- Inconsistency. The same prompt may yield different results each time.
How do you get better results?
The prompt shapes the output. Be specific about the task, audience, format, and constraints, and give an example when you can. Our guide on how to write good prompts covers this in detail.
Is generative AI the same as a chatbot?
No. A chatbot is one application of generative AI, specifically a conversational interface. Generative AI also powers image tools, code assistants, and more. See what is an AI chatbot. For more reading, try Wikipedia on generative AI.
Generative AI in Based
Based uses generative AI for chat replies, summaries, and images. Use /imagine followed by a description to generate an image, or mention Based to chat. Every slash command also works with the "." prefix, such as .imagine. Details are on the image generation docs and the summaries feature page.
Frequently asked questions
What is generative AI in simple terms?
It is AI that makes new content, like text, pictures, or code, from a description you give it. It builds the output from patterns learned in training data.
What is the difference between generative AI and AI?
AI is the broad field. Generative AI is the part focused on creating new content rather than only classifying or predicting.
Is generative AI always accurate?
No. It can produce wrong or invented details that sound convincing, so check anything important.
What are examples of generative AI?
Chatbots that write replies, tools that turn text into images, systems that generate speech or music, and assistants that write code.
Keep learning
- AI basics
What is artificial intelligence (AI)?
Artificial intelligence is software that performs tasks that normally need human judgment, like understanding language or spotting patterns. Here is how it works.
- AI basics
What is a large language model (LLM)?
A large language model is an AI trained on huge amounts of text to predict and generate language. Learn how LLMs work, what they do well and where they fail.
- AI basics
What is an AI chatbot?
An AI chatbot is software you talk to in everyday language, powered by a language model. Learn how it works, how it differs from older bots, and its limits.
- Voice and media
What is AI image generation?
AI image generation creates new images from text prompts using models like diffusion. Learn how it works, how to write good prompts, and its limits.
- AI basics
How to write good AI prompts
A good AI prompt states the task, context, format and constraints clearly. Learn practical steps, examples and common mistakes to get better answers.
Try it in your server
Based is an AI Discord bot in 300+ servers. Every slash command also works with the "." prefix.