Artificial intelligence keeps changing fast, and Drupal keeps pace with it. Drupal AI modules are the contributed modules built on the AI (Artificial Intelligence) framework that connect Drupal to large language models, enabling content automation, semantic search, and agent-based site configuration. What started as a handful of experimental modules in 2025 is now a funded, staffed initiative with agents, governance tooling, and a growing provider ecosystem.

In this guide, we will walk you through the Drupal AI modules worth knowing today and onwards, starting from the framework that powers all of them to the agents that can build configuration on their own. We have built several AI modules on top of this same framework ourselves, so this list also reflects what we've learned building with it.

Key Takeaways:

• The AI (Artificial Intelligence) module is the framework behind Drupal AI. It runs on over 17,000 sites and connects to 48+ providers, including OpenAI, Anthropic, Google Gemini, and AWS Bedrock.
• AI Automators and AI CKEditor deliver value fastest. Both are stable and need no custom code.
• AI Agents can create content types, fields, and taxonomy from a prompt. Drupal's 2026 roadmap builds governance around that, including approval steps and audit trails.
• You can run Drupal AI entirely on your own infrastructure with providers like Ollama. No content has to leave your servers.
• The Drupal AI Initiative has 28 supporting organizations and $1.5M+ in committed funding.

The AI framework behind every module on this list

Almost everything on this list sits on top of one module: AI (Artificial Intelligence). It works as the plumbing that lets Drupal talk to any AI service through one shared, consistent interface, so developers write against one API instead of rewriting integrations for each vendor.

As of August 2026, the module reports more than 17,000 active installs and sits at stable release 1.4.6. It abstracts 48+ AI providers, including OpenAI (ChatGPT, DALL·E), Anthropic (Claude), Google Gemini, AWS Bedrock, Azure AI, and self-hosted options you run yourself.

Think of it in three layers. The AI module is the framework: it defines shared settings, logging, and operation types like text generation, image generation, and embeddings. Provider modules are the connectors, each one linking the framework to a specific AI service. Submodules and contrib modules are the consumers: AI Automators, AI CKEditor, AI Agents, and AI Search all sit on top of the framework and use whichever provider you configure.

Swap a provider later, and you change a setting. You do not rewrite the integration.

The 16 Drupal AI modules to know in 2026

Not every module here is ready for the same job. Some are stable enough to deploy today. Others deserve a pilot first, and a few are worth watching but not building on yet. The list below groups them by what they do, but the numbering runs straight through all 16. A quick-reference table with deploy status sits near the end if you want the short version first.

Framework

1. AI (Artificial Intelligence):

The core framework covered above, and the one every other module on this list depends on. Installing it gives you the shared configuration screens, the operation types like chat, text generation, image generation, and embeddings, and the plugin system providers hook into. Install this first, alongside a provider module for whichever service you want to use. 

Why install it:

• Install this module first, before any other module on the list; everything else depends on the plugin system and provider connections it sets up.
• Skip it, and none of the other fifteen modules here have anything to plug into.

Editorial and content automation

2. AI Automators:

This submodule fills in fields automatically when content is saved: a summary generated from a body field, tags pulled from an article, alt text written from an image. For a content-heavy site, this is often the highest-return module in the whole ecosystem. It needs no custom code and no separate interface. Editors keep working in Drupal exactly as they do today, and the fields fill themselves in behind the scenes.

Why install it:

• It removes the busywork of writing summaries, pulling tags, or drafting alt text by hand on every save.
• This needs no custom code or separate interface, so editors keep working in Drupal exactly as they do today.

3. AI CKEditor:

AI writing assistance built directly into the content editor, not a separate tool editors have to switch to. Highlight any text and the toolbar offers a set of everyday editorial actions: rewrite, translate, summarize, adjust tone, fix spelling, reformat HTML, or run a custom prompt of your own, all applied in place without leaving the content form. 

Why install it:

• AI CKEditor keeps writing quality consistent across multiple authors without adding a separate review step.
• It puts rewrite, tone, translation, and summarization tools right in the toolbar editors already use.

4. AI Image Alt Text:

Generates image alt text automatically, which matters for both accessibility compliance and image search rankings on media-heavy sites. AI Automators can do this same job as a field-level rule if you already run that module; Automators is the better choice when you want alt text generated on save across multiple image fields. Either way, you will need a vision-capable model behind it.

Why install it:

• It closes accessibility gaps on media-heavy sites without a manual tagging project.
• It also clears a backlog of untagged images far faster than doing it node by node.

SEO and structured content

5. Drupal SEO Studio: AI Meta Tags:

Writing meta tags manually across a large site is tedious. This module drafts optimized meta titles and descriptions, keeping content aligned with current search algorithms and freeing up editorial time.

Why install it:

• This module keeps meta titles and descriptions consistent once your page count outgrows what one person can tag by hand.
• It frees up editorial time that would otherwise go into repetitive SEO fields.

6. SEO AI:

Goes a step further with on-demand SEO audits. It reviews content and metadata, then suggests concrete fixes, giving teams one AI-driven analysis tool instead of several external services to juggle.

Why install it:

• SEO AI replaces two or three separate SEO tools with one AI-driven audit inside Drupal.
• It surfaces concrete fixes instead of just flagging problems, so there is less guesswork after the audit.

7. AI Schema Markup Generator:

Runs on the same AI framework and reads a node's field data to generate valid Schema.org JSON-LD automatically. It solves a problem that scales badly by hand: keeping structured data accurate across hundreds of content types instead of writing it per node. It needs an OpenAI connection to run.

Why install it:

• It moves structured data updates out of the developer's backlog and into the content-save workflow.
• It also keeps Schema.org JSON-LD accurate across hundreds of content types without writing it per node.

8. CKEditor 5 Premium Features:

Brings real-time collaboration, track changes, and AI writing assistance into the editor itself. For teams managing multiple authors, it cuts review cycles without leaving Drupal. It ships as a paid subscription add-on.

Why install it:

• CKEditor 5 cuts review cycles for teams with more than a handful of contributors.
• It adds real-time collaboration and track changes without leaving Drupal for another tool.

9. AI Readme Generator:

Writes README.md files for custom and contrib modules based on code and structure, keeping documentation consistent across development teams without manual write-ups. Point it at a module, and it drafts the file directly from what is already there, so a developer's job becomes reviewing and tightening the wording rather than starting from a blank page. 

Why install it:

• This module keeps documentation from falling behind as a codebase grows past a few custom modules.
• It turns README writing into a review-and-edit task instead of a blank-page one.

Agents and search

10. AI Agents:

This is where Drupal AI moved from answering questions to taking action. An agent is not a chatbot that talks about your site. It is a tool that can change it. AI Agents ships with three built-in agents: a Field Type Agent that creates and edits fields, a Content Type Agent that manages node types, and a Taxonomy Agent that handles vocabularies and terms. Ask one to add a "reading time" field to your article content type, and it does the work directly. The module also includes an Agents Explorer for testing agent behavior before you trust it in production, and a Modeler API for building agent logic visually.

Why install it:

• It hands off small, repetitive configuration requests, like a new field or taxonomy term, that would otherwise sit in a developer's queue.
• This module comes with an Agents Explorer, so you can test agent behavior before trusting it in production.

Agents that edit configuration raise an obvious question: who is watching them? Drupal's 2026 AI roadmap lists advanced governance as one of its eight core capabilities, including batch approvals, branch-based versioning, and audit trails for AI-driven changes. That tooling is still maturing, so treat AI Agents as pilot-stage and test configuration changes on staging before letting an agent touch production.

11. AI Search:

Adds semantic search over your site content, and can power retrieval-augmented generation (RAG), where a chatbot answers questions using your own content instead of guessing. It needs a vector database to store content embeddings. Drupal supports several, including MariaDB, which runs vector search natively without adding an external service, and PostgreSQL with the pgvector extension. Pinecone, Milvus, and Azure AI Search are also supported for teams that prefer a managed vector store.

Why install it:

• This module picks up user intent that exact-match keyword search misses on a growing site.
• It can power retrieval-augmented generation (RAG), so a chatbot can answer from your own content instead of guessing.

12. Chatbot API:

A framework for integrating chatbots with platforms like Slack, Messenger, and custom surfaces. It remains useful for teams that need that specific integration surface, though AI Agents now cover action-taking chat flows that Chatbot API was never designed for. Worth evaluating against newer AI Agents chat flows before you build on it.

Why install it:

• The module is worth it only when you need a chat surface, like Slack or Messenger, that lives outside Drupal itself.
• It is still useful for that specific integration even though AI Agents now cover most action-taking chat flows.

Legacy

13. ChatGPT Content Assistant:

Integrates ChatGPT directly into the Drupal UI for drafting articles, product copy, and keyword extraction. It predates the AI framework. For new builds, AI CKEditor paired with a provider module now covers most of the same ground with tighter framework integration. Keep it in mind if you already run it, but weigh it against the newer path before adding it fresh.

Why install it:

• It makes sense to keep it if it is already part of your workflow; ripping it out is not necessary.
• It’s not worth adding fresh: AI CKEditor with a provider module covers the same ground with tighter integration.

Governance and watch list

14. Context Control Center:

Currently in Beta 4, released August 16, 2026. It lets a site ground AI responses in its own brand guidelines, compliance rules, and style standards, and controls which agents can see which context. This is the governance layer that decides what an agent is allowed to know, beyond what it is allowed to do. Pilot it rather than deploying it site-wide.

Why install it:

• It keeps agents from acting on brand guidelines or compliance rules they were never meant to see.
• It’s worth setting up before you scale agent use across the site, not after.

15. AI Observability & Guardrails:

Logs prompts, tracks usage, and enforces safety rules, testing agent behavior before it reaches production. Expect this to mature alongside the governance capabilities on the 2026 roadmap. 

Why install it:

• It gives you a record of what got generated and from which prompt, especially when more than one editor is using AI features.
• The module enforces safety rules and tests agent behavior before changes reach production.

16. ECA + AI / Agents Modeler API:

ECA (Event-Condition-Action) is Drupal's general-purpose automation module: something happens on the site, like a node being saved, conditions get checked, and an action runs, all wired together visually instead of in code. Its AI integration lets that action be an AI call, such as generating a short SEO summary the moment an article is saved with an empty summary field. The Agents Modeler API applies that same visual, no-code approach specifically to configuring how AI Agents behave. 

Why install it:

• It ties AI actions to site events beyond content fields, like a new user registering or a scheduled cron run.
• The module also allows you to configure agent behavior visually through the Agents Modeler API, without writing code.

One more tool worth a mention, even though it is not AI: Moderated Content Bulk Publish lets you bulk-publish or unpublish moderated content from the admin panel, with content pinning and quick language switching built in. It is not part of the sixteen above, but it solves a real editorial bottleneck that sits right next to the ones these AI modules address.

Why install it:

• The module is handy on any site where content sits in moderation queues in batches rather than one at a time.

Running Drupal AI without sending data to a third party

You only send data to a hosted provider if you choose to; the framework does not lock you into any single vendor. With a hosted provider like OpenAI or Anthropic, the content you send in a prompt leaves your infrastructure to reach that provider's API. Review each provider's data handling terms before you configure it, and check what the AI module logs locally through its observability tooling.

If that is not an option for your compliance requirements, providers like Ollama let you run models on your own infrastructure, so prompts and responses never leave your network. Teams evaluating self-hosted options also look at vLLM, Mistral, and Hugging Face-hosted models, depending on what hardware and licensing fit their setup. This is usually the deciding factor for public sector, healthcare, and finance teams that cannot send content to a third-party API. The tradeoff is managing your own model infrastructure instead of paying per token.

What is coming next for Drupal AI?

The Drupal AI Initiative launched in June 2025 and has scaled fast since. It now counts 28 supporting organizations, more than 50 contributors, and over 23 full-time-equivalent contributors pledged collectively, backed by more than $1.5 million in combined cash and in-kind contributions. Two dedicated workstreams, Innovation and Product Development, split the delivery work under contracts that began in January 2026.

We are one of those 28 organizations. Specbee sponsors the Drupal AI Initiative so AI capability is built into Drupal itself, along with security and governance. Our client platforms inherit that work. With our AI-orchestrated framework, we're able to deliver Drupal projects faster with senior engineers governing every step. We build AI features into client sites: semantic search, content generation, editorial workflows, and SmartLinker AI, our module for internal linking inside CKEditor.

The 2026 roadmap groups its work into eight capabilities: page generation, context management, background agents, design system integration, content creation and discovery, advanced governance, intelligent website improvements, and multi-channel campaigns.

Two modules from the list above are worth watching most closely as this roadmap plays out: Context Control Center, which governs what agents can see, and AI Observability & Guardrails, which governs what they log and how their behavior gets checked before it reaches production.

Quick reference: all 16 modules at a glance

Bonus, not AI-specific: Moderated Content Bulk Publish.

Check out the full lineup of Drupal AI modules here for anything not covered above. 

Final thoughts

Drupal AI stopped being a side project somewhere in the last year. It now has a funded initiative, a stable framework running on tens of thousands of sites, and modules that handle real editorial work without custom code.

The gap left to close is not capability. It is judgment: knowing which modules to run today, which to pilot on staging, and which to leave on the watch list until governance catches up. That is the call this guide was built to help you make. But this list is just the tip of the iceberg. Check out the full lineup of Drupal AI modules here.

If you want help choosing the right modules for your stack or building on the AI framework, talk to us. We've built on it ourselves.

Frequently Asked Questions

Which Drupal AI module should I install first? 

Install the AI (Artificial Intelligence) module plus one provider module. It is the framework everything else builds on, and it runs on more than 17,000 sites. Add AI CKEditor next if your team writes content in Drupal, or AI Automators if you want fields to populate themselves on save. Both are stable and need no custom code.

Does Drupal AI send my content to OpenAI? 

Only if you choose an OpenAI provider. The AI module abstracts 48+ providers, and several run entirely on your own infrastructure. Ollama and similar self-hosted providers keep every prompt and response inside your network. For regulated industries, this is usually the deciding factor, and it is a configuration choice, not a code change.

What is the difference between the AI module and an AI provider module? 

The AI module is the framework. It defines how Drupal talks to any AI service and gives you shared settings, logging, and operation types. A provider module is the connector to one specific service, such as OpenAI or Anthropic. You need both. Swapping providers later means changing a setting, not rewriting your integration.

Can AI agents change my Drupal site configuration? 

Yes. The AI Agents module ships with agents for content types, fields, and taxonomy that can create configuration from a plain-language prompt. Drupal's 2026 roadmap is building governance tooling, including approval steps and audit trails, around that capability. Treat agents as pilot-stage for production sites, and test them on staging first.

How much does it cost to run AI on a Drupal website? 

Cost comes down to three factors: the provider you pick, which operations you run, and how often they fire. Embeddings cost far less per call than text generation, and a meta tag generator that runs once per node is cheaper than a chatbot answering hundreds of daily queries. A self-hosted provider like Ollama trades per-token billing for infrastructure cost, which usually wins at high volume once server upkeep is factored in.

Do Drupal AI modules work with Drupal 11 and Drupal CMS? 

Yes. The AI module's current stable release is 1.4.6, published August 5, 2026, and it supports Drupal 10 and 11. Drupal CMS ships AI capability as part of its product strategy, so newer installs get closer integration out of the box. Check each contributed module separately, since ecosystem modules track compatibility at their own pace.

What is the best Drupal module for AI-generated alt text? 

AI Image Alt Text handles this directly, and AI Automators can do the same job as a field-level rule if you already run that module. Automators is the better choice when you want alt text generated on save across multiple image fields. Both need an active provider with vision support configured.

Is Drupal AI production-ready or still experimental? 

The framework is production-ready. AI CKEditor, AI Automators, and stable provider integrations run on live sites today. The agent layer and Context Control Center are still maturing, so pilot those rather than deploying them site-wide. The gap in 2026 is governance tooling, not raw capability.