AI Engine by Jordy Meow is the most capable free AI plugin for WordPress. It connects to OpenAI (GPT-4, GPT-4o), Anthropic (Claude), Google (Gemini), and other AI providers through their APIs, bringing those models directly into the WordPress editor, the frontend as a chatbot, and your content workflow as an AI assistant. The free version on WordPress.org covers more ground than most competing plugins charge for.
What AI Engine Actually Does
AI Engine is a connector plugin – think of it as the bridge between WordPress and whichever AI model you choose – it does not run any AI itself. It sends your requests to whatever AI API you configure and returns the responses into WordPress. This means the AI quality depends entirely on which model you use, and it means you pay the AI provider (OpenAI, Anthropic) directly for API usage rather than paying a markup to a WordPress plugin. For high-volume content generation, this is significantly cheaper than plugins that include AI access in their subscription price.
The main features: a chatbot widget for your frontend (powered by your chosen AI model), an AI assistant in the post editor for writing and rewriting content, image generation from text prompts (Dall-E or Stable Diffusion via AI Engine), AI-powered forms that respond intelligently to user input, and a playground for testing prompts directly from the WordPress admin.
Step 1: Get Your API Key
AI Engine needs an API key from whichever AI provider you want to use. For OpenAI: go to platform.openai.com, create an account, go to API Keys, and generate a key. For Anthropic: console.anthropic.com, create an account, generate an API key. For Google Gemini: go to aistudio.google.com and create an API key.
Enter the key in AI Engine -> Settings -> API Keys. You can add keys for multiple providers and switch between them per feature. A common setup: use GPT-4o for content generation (better at following instructions) and Claude for the chatbot (better conversational responses). Both keys work simultaneously for different features.
Important: API usage is billed by the AI provider based on tokens processed. Set a monthly spending limit in your OpenAI or Anthropic account dashboard before generating content at scale. A GPT-4o key without a spending limit on a busy site can produce unexpected charges.
Need help integrating AI into your WordPress workflow? Describe your project and get a free estimate.
Step 2: Configure the AI Chatbot
Go to AI Engine -> Chatbots -> Add New Chatbot. Configure:
- AI model – which model this chatbot uses. GPT-4o mini is the most cost-efficient for general customer support. GPT-4o is more capable but costs significantly more per conversation.
- System prompt – the instructions that define the chatbot’s personality, knowledge scope, and behaviour. This is the most important setting. A vague system prompt produces a generic AI; a specific prompt produces a useful assistant. Example: “You are a customer support assistant for [Store Name]. You help customers with order questions, product information, and shipping. You do not discuss competitor products or pricing. If you cannot answer a question, ask the customer to contact support@storename.com.”
- Context – attach your FAQ page, product descriptions, or any content as context the chatbot can reference when answering questions.
Place the chatbot on your site using the shortcode [mwai_chatbot id="X"] or the Gutenberg block.
Step 3: AI Writing Assistant in the Post Editor
AI Engine adds an AI assistant panel to the WordPress post editor (both Gutenberg and classic editor). Select any text and ask the AI to rewrite it, expand it, summarise it, or change its tone. Generate a blog post outline from a keyword. Write a product description from bullet points. The assistant respects context – selecting a paragraph and asking for expansion gives better results than asking from a blank page.
The editor integration works differently depending on your editor. In Gutenberg, AI Engine adds a sidebar panel and toolbar options. In the classic editor, it adds an AI menu item. In both cases, the model and prompt settings from your AI Engine configuration apply.
Step 4: Content Generation Templates
AI Engine Pro adds content generation templates – pre-built prompts for common content types that you run on demand. Create a template for “Product SEO Description” that takes a product name and key features as inputs and generates an optimised description. Create a “Blog Post from Outline” template that expands a bullet-point outline into a full post. Run these templates from the AI Engine dashboard or trigger them programmatically via hooks for bulk content generation.
Setting Spending Limits to Avoid Surprise Bills
API billing without limits is a real risk. A misconfigured chatbot that loops, a bulk generation script that runs unchecked, or a caching issue that makes repeated identical requests can generate unexpected costs. Before using AI Engine in production, set hard spending limits in your AI provider dashboard.
In OpenAI: go to Settings -> Limits and set a monthly budget cap. OpenAI sends an email when you reach 80% of the limit and pauses API access at 100%. In Anthropic: set usage limits in the Console -> Settings -> Billing. In AI Engine itself, you can limit token usage per user role in Settings -> Limits – for example, allowing editors to generate up to 10,000 tokens per day to prevent runaway usage from a single user.
The practical approach for a new installation: start with a $10/month limit, monitor daily for the first two weeks, and increase the limit once you understand your actual usage pattern. A typical site with a support chatbot and a content team doing occasional AI writing rarely exceeds $5/month on GPT-4o mini.