Artificial intelligence (AI) encompasses many complex, emerging technologies that once required human input and can now be performed by a computer. Broadly speaking, AI is a non-human program or model that demonstrates a broad range of problem-solving and creativity.
The acronym AI is often used interchangeably to represent various types of technologies within the field of artificial intelligence, but these can vary greatly in scope.
There are a number of terms and concepts which define artificial intelligence and machine learning, that you may find useful. Here you'll find terms common in Chrome's documentation, most critically, around client-side AI.
Client-side AI
While most AI features on the web rely on servers, client-side AI runs in the user's browser and performs inference on the user's device. This has numerous benefits, including lower latency, reduced cost to create features, increased user privacy, and offline access.
Client-side AI relies on smaller, optimized models, that are optimized for performance. It's possible such models to outperform larger server-side models for specific tasks. Assess your use case to determine what solution is right for you.
Built-in AI

Built-in AI is a form of client-side AI, where the smaller models are built into the browser. For Chrome, this includes Gemini Nano and expert models. Once these models are downloaded, all websites and web applications that use built-in AI can skip the download time and get right to feature execution and local inference.
The built-in AI APIs are designed to run inference against the right type of model for the task. For example, the Prompt API runs inference against an LLM, while the Translator API runs inference against a built-in expert model.
Server-side AI
Server-side AI encompasses cloud-based AI services. Think Gemini 1.5 Pro running on a cloud. These models tend to be much larger and more powerful. This is especially true of large language models.
Hybrid AI
Hybrid AI refers to any solution including both a client and server component. For example:
- Client-side models that have a fallback to server-side models, built for tasks
that cannot be completed effectively on the device.
- There may be a lack of resources on the device.
- The model or API is only available in certain environments.
- A model split between client and server for security.
- For example, you could split a model such that 75% of the execution happens in the client, while the remaining 25% is performed on a server. This brings client-side benefits, while allowing part of the model to be off-device, thus remaining private.
If you use the Prompt API, you can set up hybrid architecture with Firebase AI Logic.
Generative AI
Generative AI is a form of machine learning that helps users create content that feels familiar and mimics human creation. Generative AI uses large language models to organize data and create or modify text, images, video, and audio, based on supplied context. Generative AI goes beyond pattern matching and predictions.
A large language model (LLM) has numerous (often billions) parameters that you can use to perform a wide variety of tasks, such as generating, classifying, or summarizing text or images.
Natural language processing (NLP)
Natural language processing is a class of ML that focuses on helping computers comprehend human language, from the rules of any particular language to the idiosyncrasies, dialect, and slang used by individuals.
Agent or AI agent
An agent is software that autonomously plans and executes a series of actions to complete a task on a user's behalf, while adapting to changes in its environment. Actions may include API functions or database queries, performed on a webpage or through a third-party application, such as Project Mariner.
A chatbot is not inherently an agent. While a chatbot responds to a messenger (be it human or otherwise) and relies on a model to generate content, such as answers to questions, an agent interacts with tools or a database to complete a task.
Additional resources
If you're new to AI on the web, check out our collection of web.dev AI resources.