Java and ChatGPT: Creating an AI Assistant in Code

Java Meets Artificial Intelligence: Why It Matters

The world of Java development is experiencing a real revolution. Just a few years ago, integrating artificial intelligence into Java applications seemed like the domain of large corporations with teams of data scientists. Today, however, every Java developer can add the power of language models like ChatGPT to their project. And this is not just a trendy fad — it is a real tool that can automate routine tasks, improve user experience, and open new horizons for your applications.

In this article, we will break down how to use ChatGPT in Java development. You will learn how to connect to the OpenAI API, write a simple AI assistant, and integrate it into a real project. No magic — just pure Java code and practical examples. Let's go!

What is ChatGPT for a Java Developer?

ChatGPT is a language model from OpenAI capable of understanding and generating human-like text. For a Java developer, this means the ability to add features to your application that previously required complex algorithms and large amounts of data: answering user questions, generating content, analyzing text, writing code, and much more.

The main advantage is that you don't need to train the model yourself. OpenAI provides a ready-made API that can be called from a Java application. You simply send a request (prompt) and receive a response. The neural network handles all the complex natural language processing work.

Preparation: What You Need to Get Started

Before writing code, let's prepare everything you need. You will need:

\n
• Java Development Kit (JDK) version 11 or higher (we recommend 17/21).
\n
• Build system (Maven or Gradle) — for managing dependencies.
\n
• OpenAI API key — can be obtained in your personal account after registration.
\n
• HTTP client — we will use the built-in HttpClient from Java 11.
\n

By the way, we will need a library for working with JSON. I recommend com.google.code.gson:gson. Add it to your pom.xml (for Maven):

<dependency>\n    <groupId>com.google.code.gson</groupId>\n    <artifactId>gson</artifactId>\n    <version>2.10.1</version>\n</dependency>

For Gradle:

implementation 'com.google.code.gson:gson:2.10.1'

Ready? Then let's move on.

Writing the First Java Code to Communicate with ChatGPT

Our first example is a classic chatbot. We will send a request to the gpt-3.5-turbo model and receive a response. This is the foundation on which we will build more complex solutions.

Create a class ChatGPTClient.java:

import java.net.URI;\nimport java.net.http.HttpClient;\nimport java.net.http.HttpRequest;\nimport java.net.http.HttpResponse;\nimport com.google.gson.JsonArray;\nimport com.google.gson.JsonObject;\n\npublic class ChatGPTClient {\n\n    private static final String API_URL = "https://api.openai.com/v1/chat/completions";\n    private static final String API_KEY = "YOUR_API_KEY"; // Don't forget to replace!\n\n    public static String askChatGPT(String userMessage) throws Exception {\n        HttpClient client = HttpClient.newHttpClient();\n\n        // Form the request body in JSON\n        JsonObject requestBody = new JsonObject();\n        requestBody.addProperty("model", "gpt-3.5-turbo");\n\n        JsonArray messages = new JsonArray();\n        JsonObject systemMessage = new JsonObject();\n        systemMessage.addProperty("role", "system");\n        systemMessage.addProperty("content", "You are a helpful Java assistant.");\n        messages.add(systemMessage);\n\n        JsonObject userMsg = new JsonObj