Feed aggregator

Drupal.org blog: GitLab issue migration: a contributor's perspective

Drupal Planet -

This is the fourth post in our GitLab issue migration series. The earlier posts focused on what is changing and how maintainers should set up their projects. This one is for the rest of us — the people who file bugs, review code, push fixes, and triage queues without wearing a maintainer hat. If your favorite contrib project has just moved its issues to git.drupalcode.org, here's what you need to know.

What's changed at a glance

When a project's issues are migrated, they move from www.drupal.org/project/{name}/issues to git.drupalcode.org/project/{name}/-/work_items. Old URLs redirect to the new ones, and issue numbers (NIDs) are preserved as GitLab IIDs — so an #3409678: Opt-in GitLab issues you find in a commit message will still resolve to the same issue.

In GitLab, "issues" are technically a subtype of "work items," but the term issue still applies, and you'll see it throughout the UI. If you've worked on any GitHub or GitLab project before, the experience will feel familiar.

What still works the way you're used to

A lot has not changed:

  • You can still create issues without any special role.
  • The contribution credit system is unchanged. Every comment you make on a GitLab issue still syncs to its contribution record automatically. The credit UI itself still lives on drupal.org.
  • Shared issue forks are still the way to collaborate (on code). Drupal still doesn't use personal forks; collaboration on a single fork remains the model. The fork management UI also still lives on drupal.org.
  • The familiar workflow conventions can still apply — Needs work, Needs review, RTBC, priority levels, and so on. They were migrated as scoped labels (more on this below).
  • GitLab CI still runs your tests the same way it has for the past few years.
  • Your Drupal.org login still works. Single sign-on means no extra account to manage.
  • Cross-references and parent/child relationships still exist, just with slightly different syntax (more on that below).
Some things actually got better

It's worth naming a few real wins for contributors:

  • A modern issue editor with proper markdown, real code blocks, syntax highlighting, image paste, and a mobile-friendly UI.
  • Issues and merge requests now live alongside the code, pipelines, and CI logs on git.drupalcode.org — far less context-switching during code review.
  • Much better filtering, search, and saved queries on issue lists.
  • Issue boards (Kanban-style) are available for projects that want them.
Who can edit what

A few permission details are worth knowing up front, because they're tighter than what you may be used to on the old issue queue:

  • Anyone can comment on an issue.
  • Only the original author or project maintainers can edit the issue description. This is a real change — on drupal.org, any logged-in user could edit an issue summary. If you'd like a description updated and you're not the author, leave a comment with the suggested wording so the author or a maintainer can apply it.
  • Labels and other metadata (priority, version, category, component, tags) can only be edited by users with a Planner role or higher on the GitLab project.

That last point is real friction for contributors who triage and label issues, and we're addressing it directly. #3559846: Allow changing GitLab issues labels for all contributors is building a label-management UI that will live on drupal.org, alongside the existing contribution credit and issue fork management screens. Once it ships, any contributor will be able to manage labels on any issue without needing a project role on GitLab. This is also an upstream issue, but it doesn't seem to be worked on.

Until then, if metadata needs updating, leave a comment noting what should change. Maintainers and other contributors with the role can apply it.

Old workflow conventions can still apply

Good news for anyone with muscle memory for Drupal's NW / NR / RTBC dance: the conventions weren't dropped in the migration. They were preserved as scoped labels on GitLab issues — state::rtbc, the equivalent state labels for needs-work and needs-review, priority labels, and so on. Each project's setup may vary, but the familiar conventions carried over, and contributors can keep using them.

Convention Now Needs work state::needsWork label (alternative: MR set to Draft) Needs review state::needsReview label (alternative: MR set to Ready) RTBC state::rtbc label (alternative: MR approval) Needs reroll Push a rebase to the issue fork branch

Two notes:

  • Applying these labels yourself currently requires the Planner+ role on the project, until #3559846 ships the contributor label UI. In the meantime, a comment indicating the status you'd assign is the right move.
  • Merge request states are a useful parallel signal: an MR's Draft / Ready toggle and approval status reflect the actual code change being reviewed, which often communicates more clearly than a label on the parent issue. Some projects lean heavily on MR state, some on labels, some use both — there's no single right answer.
Day-to-day: how to do common things Create an issue

Navigate to the project, click Issues in the left sidebar, then New issue. The form is just a title and description; labels and metadata are added afterwards by users with the appropriate role. If a project has set up issue templates (markdown files in the repo), you'll see them in a dropdown.

The first auto-generated comment

The first comment on every new issue is posted by DrupalBot. It's the bridge to the things that still live on drupal.org:

  • A link to the contribution record (where credit is tracked).
  • A link to the fork management page, where you can create or request access to a shared fork, view existing MRs, and open a new one.
  • Once #3559846 ships, a link to the contributor label management UI.
Forks and merge requests

The fork management screen on drupal.org works the same way for GitLab issues as it has for Drupal.org issues. From there, you can create a shared issue fork, request access if one already exists, push a branch, and open an MR. Branching and merging happen in GitLab's native UI, where they're already optimized.

Linking issues across systems

During the transition, contributors will be working with both Drupal.org issues and GitLab issues, sometimes in the same comment. The syntax differs by direction:

  • Drupal.org → Drupal.org issue Brackets around the issue number (without spaces): [ #123456 ] (unchanged)
  • GitLab → GitLab issue (same project): #123456 (no brackets)
  • Cross-platform (either direction): paste the full URL

For "related issues" entries on Drupal.org, always use the full URL when pointing at a GitLab issue.

Things you might notice on migrated issues

A few oddities are worth flagging if you're working through historical issues:

  • DrupalBot is listed as the author on every migrated issue. (Issues created after migration correctly show their actual creator.) The GitLab API forced a tradeoff between preserving the original author and preserving the issue ID. We chose to preserve the ID so that #123456 still maps to the same issue. The original author's name is preserved in the first line of the issue description.
  • Old drupal.org issue URLs redirect to their GitLab equivalents, so existing links in commit messages, blog posts, and external references continue to work.
Reporting bugs and getting help

Found a bug in the migration itself or in the integration between Drupal.org and GitLab? Please file it in the Drupal.org customizations issue queue.

Have a question, or want to share feedback on the new workflow? Join the #gitlab-issues-feedback channel on the Drupal community Slack.

We're actively iterating on this transition based on what we hear from contributors and maintainers in opted-in projects. The more feedback we get now — while we're still in the opt-in phase — the better the experience will be when the rest of contrib gets batch-migrated.

Reference: GitLab documentation

For more detail on any of the GitLab features mentioned in this post, the official GitLab documentation is the canonical source.

Issues and work items

Labels and permissions

Merge requests

Markdown

Related blog posts in this series:

Related issues

* We used Claude AI to refine our first draft and help link related materials like the GitLab documentation.

Drupal AI Initiative: From Leuven to Athens: Celebrating One Year Since the Drupal AI Initiative Took Shape

Drupal Planet -

One year ago, at Drupal Developer Days in Leuven, something special happened.

The Drupal AI Initiative was not officially launched yet. That would happen later, in June. But Leuven was where the spark happened. It was where the first real momentum came together. Where conversations turned into commitment. Where a shared belief became a shared plan.

Five companies stepped up to kickstart the initiative: Dropsolid, Acquia, 1xINTERNET, FreelyGive, and Salsa Digital. Together, they helped turn an ambitious idea into the beginning of a movement.

Now, one year later, as we gather again at Drupal Developer Days in Athens, we celebrate one year since that moment of conception.

Leuven was where the initiative was kickstarted. June was when it officially went live. Athens is where we celebrate how far it has come.

A year of momentum, collaboration, and delivery

The Drupal AI Initiative was created with a bold ambition: to help Drupal become the leading open source CMS for AI-powered digital experiences.

But from the beginning, this was never just about adding AI features.

It was about building AI into Drupal in a way that reflects the values of the Drupal community: open, flexible, responsible, transparent, and collaborative. It was about giving organizations the tools to innovate with AI while keeping control over governance, content, security, editorial workflows, and long-term digital strategy.

Over the past year, the initiative has grown from a spark in Leuven into one of the most ambitious collaborative efforts in Drupal’s history.

What has been delivered?

Since the official launch of the Drupal AI Initiative, the team has made major progress. The amount of installs is growing significantly, 13980 at the time of writing. Adoption is accelerating. According to shared data we’re growing at about 260 sites per week and accelerating.

This is only the sites that share numbers, the real share is much higher. 

Growing numbers of Drupal AI Partners

Between Drupal Con Vienna and Chicago, the initiative added 12 new partners, a total of 34, representing a 50% increase in participation. We are on track to match this growth in support between now and DrupalCon Rotterdam, a key goal for this year.

Delivering capabilities at scale

The initiative also successfully established and executed the delivery management RFP process, putting important operational frameworks in place, including:

  • Partner expectations documentation
  • Onboarding processes
  • Development sprint processes
  • Contribution tracking
  • Team planning sheets
  • Weekly status reporting

These may sound like operational details, but they are what make collaboration at scale possible. They help turn enthusiasm into structure, and structure into delivery.

The Drupal AI Initiative has become the largest multi-company collaboration in Drupal community history.

- Dries Buytaert

The initiative is now actively funding critical roles across multiple organizations, including product management, innovation management, technical leadership, and program management. This marks a major milestone: the Drupal AI Initiative has become the largest multi-company collaboration in Drupal community history.

The 2026 roadmap was finalised earlier this year, informed by customer demand and industry insight. Delivery is actively underway.

At the same time, marketing efforts have been elevated to position Drupal as the leading AI-powered open source CMS globally, supported by ongoing storytelling and visibility through the Drupal AI Initiative blog.

Significant rise contribution

Focused effort on strategically important features, combined with a growing number of partners committing resources and strong community participation, has driven a significant increase in momentum and impact.

The tag clouds below visually represent the many Drupal community members who in the past 12 months have contributed to the AI Initiative (sized according to number of fixed issues worked on). Includes code and non code contributions. 

The following organizations have also contributed to the Drupal AI Initiative in the past year.

Marketing Drupal AI

From small beginnings with Paul Johnson and Frederik Wouters taking on marketing, we now have a cross disciplinary high performing team with 10 leads across areas of specialization.

Focussing on introducing Drupal to new audiences the emphasis has been on webinars, participating in external events and organising major new customer facing events of our own. These include Drupal AI Summit Paris and New York City (14th May 2026), The AI Summit London (10-11 June 2026) and the latest Enterprise AI Summit Rotterdam (28 September 2026).

Our work has included facilitating Southwark Council, London, winning Digital Leaders AI Impact Award 2026, producing video case studies and highlighting major new AI features announced during the DriesNote with social video content. All these activities have substantially raised Drupal’s profile to a wider audience.

Building toward Rotterdam

The next major milestone is already taking shape.

In Rotterdam, the initiative will launch an exclusive Drupal Enterprise AI event, available only to Drupal AI Initiative partners. The event will bring together European decision-makers aboard the SS Rotterdam for peer networking, customer case studies, and strategic conversations about building AI-powered content management solutions with Drupal.

Participation in this event is limited to partners who join the Drupal AI Initiative by June 30.

That creates a powerful moment for companies that want to be part of Drupal’s AI future. The initiative is scaling, the roadmap is active, the team is growing, and the opportunity to help shape what comes next is open now.

A strong foundation for what comes next

The Drupal AI Initiative is in a strong position.

With $380,000 in cash and $1.5 million in in-kind contributions, more than 50 contributors from partners, the initiative has the resources and commitment needed to continue scaling. The plan is to onboard an additional 12 partners by Rotterdam, further strengthening the team and accelerating delivery.

The message is clear: You counted on Drupal AI, and we delivered. Now we want to create more efficiency and scale.

That is what this next phase is about. More delivery. More visibility. More impact.

Thank you to the people who made this possible

This milestone belongs to many people.

It belongs to everyone who joined those early conversations in Leuven.

It belongs to Frederik Wouters, who brought the right people together at the right moment and helped create the spark that started it all.

It belongs to the five companies that kickstarted the initiative: Dropsolid, Acquia, 1xINTERNET, FreelyGive, and Salsa Digital.

It belongs to every partner, contributor, sponsor, strategist, developer, product thinker, marketer, and community member who has helped move this initiative forward.

And it belongs to the wider Drupal community, whose openness and willingness to collaborate make initiatives like this possible.

You can still join

The Drupal AI Initiative is growing, and companies can still become part of it.

If your organization believes in the future of Drupal, if you want to help shape responsible AI in open source, or if you want to be part of the group building the next generation of AI-powered content management, now is the time to join.

Become part of the 34 makers already helping to build Drupal’s AI future.

To join the Drupal AI Initiative as an organization and become a partner, contact Dominique at dominique@dropsolid.com.

From Leuven to Athens, this has been an incredible first year.

And the best part? We are only just getting started!
 

File attachments:  54456328639_3949c293a3_k.jpg adoption.png partners.png tagcommunitytop.png tagcommunitybottom.png tagcompanies.png 10team.png events.png rotterdam.png scaling.png

Centarro: How to Know If Your eCommerce Developer is Failing You

Drupal Planet -

A business spends hundreds of thousands of dollars with a developer or agency to build an eCommerce website, endures years of instability and missed deadlines, and then concludes that the platform just doesn’t work. They start eyeing Shopify or whatever choice platform the first consultant they engage recommends, hoping the grass will be greener. Meanwhile, the actual issue—an underqualified or negligent service provider—walks away unexamined.

Developer problems are often disguised as platform problems. We’ve seen this situation many times with Drupal Commerce implementations that aren’t performing as desired. We’ve even solved issues merchants put up with for years in a matter of hours. It’s not that we’re special, though we do know our own platform better than anyone else. We believe any competent Drupal developer would also be able to identify and solve these issues, possibly just as quickly.

So how do you tell the difference? How do you know your issues stem from your developer, and not your platform?

Below, we’ll give you the language and the lens to evaluate whether your developer is actually serving you well, or whether they’re the reason your Drupal Commerce site feels like it's held together with duct tape and bubblegum.

How some developers get in over their heads

A company needs a Drupal website with eCommerce capabilities, so they search for a Drupal developer. Maybe they already have a Drupal website and want to add some commerce features. Either way, they find a freelancer who has built blogs, nonprofit sites, and maybe a university portal with some advanced functionality. That person says, "Sure, I can handle commerce. It’s just another module." For a basic eCommerce website with minimal traffic, maybe they can.

Read more

Freelock Blog: When Views meets Drupal Canvas -- getting dynamic content into your Canvas page

Drupal Planet -

When Views meets Drupal Canvas -- getting dynamic content into your Canvas page John Locke Tue, 04/21/2026 - 08:00

From early days, "views" has been the killer feature of Drupal. Views is a powerful querying tool built into Drupal that allows dynamic lists and displays of content to be created without writing custom code.

Dev Corner

Jacob Rockowitz: Drupal (AI) Playground: Training and practicing building a module using AI

Drupal Planet -

Successes and failures

I am continually experiencing both successes and failures while playing in my Drupal (AI) playground. My failures usually come from expecting too much of an AI, especially when I ask it to do too many things in a single prompt. My successes with AI come when I keep things useful, simple, and achievable.

Building something useful, simple, and achievable with AI

As I've learned about and maintained new ecosystems in Drupal, I like to review all available plugins. For the Webform module, I created reports for elements, handlers, variants, and exporters. For ECA, I developed an ECA Report module. For the Meta Tag module, I contributed a patch to get a Meta Tag plugin report committed. I think having a way to browse a module's or ecosystem's plugins helps developers understand what tools are available. A Drush command for exporting plugin definitions could be used by both humans and AI.

In the past, creating and maintaining a report could be time-consuming. The new reality is that AI makes it easier to build and maintain simple things like reports. One of the most common anecdotes I hear from non-technical people who "vibe code" is that they are building websites or reports to display information.

My goal was to create a report that lists all plugin managers, plugin definitions, and individual plugin details.

There ain’t nothing fancy here

The Plugin Report module I created with AI is nothing special. Claude Code’s only challenge was getting the PHP introspection code to pass PHPStan’s level 6 coding standards. In many ways, this module served as an exercise to reinforce my ability to guide an AI in the right direction. My biggest...Read More

Très Bien Blog: Proposal for an LLM policy for Drupal Core contribution

Drupal Planet -

Proposal for an LLM policy for Drupal Core contribution

I've been following and participating in the conversation about applying AI tools to the Drupal core issue queue, and the broader community. I've been listening, reading, and experimenting quite a bit in and out of Drupal. It's been a wild ride since last December and for the past few weeks a few things started to solidify.

theodore April 21, 2026

Talking Drupal: Talking Drupal #549 - Catching up with the DDEV Team

Drupal Planet -

In Episode 549, Randy Fay and Stas Zhuk join us to discuss what DDEV is, recent improvements, and where it's headed. Module of the week is the DDEV Drupal Contrib add-on. Randy and Stas discuss priorities like reliability, consistent UX, add-ons discoverability, and new features including revamped ddev share with Cloudflare and rootless Podman support. They also cover coder.ddev.com, a cloud-based DDEV environment built on coder.com for easier onboarding and contribution, plus sustainability, community support, and challenges such as AI-driven PR volume and Stas's development constraints in Ukraine.

For show notes visit: https://www.talkingDrupal.com/549

Topics
  • Module of the Week - DDEV Drupal Contrib
  • DDev Drupal Contri Overview
  • Contrib Workflow Q&A
  • Drush in Core Debate
  • Add-on Registry and Contact
  • Drupal AI Summit Plug
  • What Is DDev
  • Stas Origin Story
  • Recent Releases and Priorities
  • DDev Share and Podman
  • Developer Experience Changes
  • Database Upgrade Pain Points
  • Coder DDev Cloud IDE
  • Cloud DDEV Basics
  • VS Code Remote Workflow
  • Pair Programming Training Wins
  • Docker Desktop Alternatives
  • Onboarding Teams Faster
  • Windows Support Reality
  • Building Through War
  • Roadmap Env File Fixes
  • Beyond Drupal Adoption
  • Addons Discovery Tools
  • Funding Community Health
  • AI Pull Requests Pressure
  • AI Agents MCP Plans
  • How To Get Involved
Resources

DDEV - https://ddev.com/ DDEV Add-on Registry - https://addons.ddev.com/ Introducing coder.ddev.com: DDEV in the Cloud - https://ddev.com/blog/coder-ddev-com-announcement/ About Stas Zhuk - https://ddev.com/blog/introducing-maintainer-stas/ Power Through Blackouts: How DDEV Community Helped Me in Ukraine - https://ddev.com/blog/power-through-blackouts-ddev-community-support/ Drush command in core - https://www.drupal.org/project/drupal/issues/3453474 Drush's Final Act - https://weitzman.github.io/blog/drush-final-act coder.com - https://coder.com/ Service hosting coder.ddev.com - https://www.hetzner.com/ Funding DDEV - https://ddev.com/blog/sustainability-for-ddev/ Gen AI DDEV newsletter note - https://ddev.com/blog/ddev-march-2026-newsletter/ Sharing Coder.ddev.com workspaces - https://github.com/ddev/coder-ddev/issues/80

Guests

Stas Zhuk - stasadev Randy Fay - ddev.com rfay

Hosts

Nic Laflin - nLighteneddevelopment.com nicxvan John Picozzi - epam.com johnpicozzi Rod Martin - DrupalHelps.com imrodmartin

Module of the Week

with Martin Anderson-Clutz - mandclu.com mandclu

DDEV Drupal Contrib - DDEV integration for developing Drupal contrib projects. As a general philosophy, your contributed module/theme is the center of the universe.

The Drop Times: Sovereignty Expires; Licences Don’t

Drupal Planet -

Europe is finally getting serious about digital sovereignty, and getting it half right. The instinct to “Buy European” is sound, but the frameworks being built around it are solving for the wrong variable. Ownership and headquarters are snapshots; they tell you where power sits today, not where it will sit after the next acquisition. Skype had every European credential imaginable. Microsoft shut it down in 2025.

The missing piece is durability. Dries and Nicholas argue, convincingly, that a sovereignty score without an open-source licensing requirement is a sovereignty score with an expiry date. The GPL licence did not stop Oracle from acquiring Sun Microsystems, but it ensured that MySQL could not be discontinued. MariaDB exists today because someone had the legal right to fork before the deal closed. That right is structural; it does not depend on which flag flies over the headquarters.

The forthcoming Cloud and AI Development Act is the real test. Europe can use it to define what makes sovereignty resilient: open licensing as a hard gate for mission-critical procurement, and supply chain assessments that distinguish between dependencies that can be replaced quickly and those that would take years to rebuild. Anything short of that risks becoming a checklist rather than a strategy.

With that, here are the key stories from the past week.

DISCOVER DRUPALEVENTORGANIZATION NEWSDRUPAL COMMUNITYSECURITY

Additional developments from across the Drupal ecosystem were published during the week. Readers can follow The Drop Times on LinkedIn, Twitter, Bluesky, and Facebook for ongoing updates. The publication is also active on Drupal Slack in the #thedroptimes channel.

Alka Elizabeth
Sub-editor
The Drop Times

Drupal AI Initiative: Drupal Is All In on AI. Now Comes the Hard Part

Drupal Planet -

Original article posted by Christoph Breidert on 1xINTERNET website

Over a decade ago, I co-founded 1xINTERNET on the conviction that Drupal was the best platform for ambitious web applications. That bet paid off. But recently, as AI began disrupting our industry, I found myself facing an unfamiliar feeling: uncertainty. For the first time in my career, the path forward wasn't entirely clear.

If you are a decision-maker navigating this shift, you likely feel the same way. We are all trying to figure out how to leverage AI's huge potential without compromising enterprise security, compliance, or content quality.

The good news is that while the broader AI landscape remains turbulent, the direction for content management systems is becoming clear.

Christoph Breidert


Christoph Breidert facilitating a Drupal AI workshop at DrupalCon Chicago 2026.

When the Drupal AI Initiative was founded in June 2025 by 1xINTERNET, Acquia, DropSolid, FreelyGive, and Salsa Digital, our mission was to chart that exact path. Today, alongside Niels Aers, my role is to manage the AI product direction so that organizations can confidently bring AI into production.

Since the founding, over 30 leading companies have joined the initiative. But a defining moment happened recently at DrupalCon Chicago 2026. During his keynote - the "Driesnote" - Drupal founder Dries Buytaert bluntly asked the community regarding the AI shift: Are you in or are you out?

The undeniable energy from the community and the rapidly intensifying momentum proved one thing: Drupal is all in on AI.

But what does "all in" actually mean? We aren't just talking about adding superficial features like chatbots or simple text generators. We have built a powerful agentic infrastructure natively into Drupal. This provides us with a robust foundation, allowing organizations to build complex AI applications and deploy autonomous agents capable of executing multi-step workflows on their behalf.

What an Agentic CMS actually requires

Let’s be clear: Agentic AI delivers incredible velocity, and every organization from SMEs to global enterprises needs that speed. But deploying autonomous agents without control is a liability. You need AI infrastructure that accelerates your workflows while ensuring that this speed doesn't destroy your content quality or violate your compliance rules.

This requires a robust governance foundation to run the infrastructure safely. The Drupal AI Initiative has spent the past months building exactly that. These are the final pieces we have built to complete the production-ready foundation:

  • AI Guardrails: Configurable rules that intercept both outgoing requests and incoming AI responses. Whether it's preventing the exposure of personal data (PII), ensuring prompt safety, or mitigating legal liability, guardrails keep the AI agents within defined boundaries.
  • AI Observability: Complete transparency into what your AI agents are doing. Every prompt, token usage metric, and model response is logged, providing a clear audit trail for compliance and cost optimization.
  • Context Control Center: AI models are useless without context. This system acts as a router, intelligently feeding the right organizational data (and only the right data) to the LLM based on the user's specific task.
Introducing AI Content Reviews

Let’s separate the hype from reality: The core foundation of Drupal AI is production-ready today. With a secure governance infrastructure now in place, we are shifting from building the engine to delivering the applications. We are shipping out-of-the-box features so organizations can immediately benefit without building complex workflows from scratch.

The first major capability rolling out is AI Content Reviews. This is not a future roadmap concept, it is a real, tangible feature designed to close the quality gap for large websites by acting as a continuous, background quality assurance partner.

It provides scalable, AI-assisted content governance that integrates naturally into how editors already work. The system evaluates content against your organization's specific rules, such as brand voice, legal compliance, SEO, and accessibility. It flags issues, explains them in plain language, and proposes concrete fixes. Crucially, human oversight remains the starting point: an editor simply reviews the flagged issues and can apply the suggested fixes with a single click.


AI Review Management Overview

Upcoming features

AI Content Reviews is just the first application of our agentic infrastructure. Following close behind is AI-powered semantic search with synthesized summaries. This allows visitors to find what they need through meaning rather than keywords, enabling the site to surface direct answers instead of just a list of results. 

We are also actively packaging AI assistants embedded natively across editorial workflows, site-building, and end-user interfaces. These capabilities have been thoroughly explored and validated in our innovation workstream and are now being readied for production use.

Want to see the full picture of what we are building? You can explore the complete Drupal AI Roadmap to see exactly where the initiative is heading next.


Overview Drupal AI Roadmap 2026.

Drupal’s architectural advantage

Why build this directly into Drupal instead of relying on external AI services or other CMS platforms? It comes down to a fundamental technological advantage. Many modern CMS platforms, especially closed SaaS products and pure headless systems, force you to rely on disconnected external API wrappers to communicate with AI. This architectural limitation means your developers have to manually rebuild your existing user permissions, workflows, and access rules in a separate middleware layer just to keep the AI secure.

Drupal AI has a distinct head start because of its deep internal architecture:

  • Co-location with the Content Graph: AI models are only as good as the context they can access. By embedding AI orchestration directly within Drupal, the AI has native, zero-latency access to your entire structured content graph. There is no integration friction.
  • Native Permissions & Access Control: Because Drupal's entity system and field-level access controls are so deeply integrated, the AI operates entirely within your existing permissions. It cannot expose, analyze, or modify content the user shouldn't see.
  • Provider-Agnostic Abstraction: Similar to what makes frameworks like LangChain powerful, Drupal AI abstracts the LLM providers (OpenAI, Anthropic, local models, etc.). But unlike external middle-tiers, Drupal enforces strict schema typing before data ever hits your database, ensuring structural integrity.

An Unmatched Ecosystem for AI Agents: Autonomous agents need tools to interact with the outside world. Because Drupal already possesses a massive, deeply established ecosystem of enterprise integrations, your AI agents can directly interact with your CRMs, ERPs, and marketing platforms. You don’t have to build custom API connectors for your AI to take action across your broader tech stack.

Moving forward

The uncertainty of the AI era remains, no one knows exactly what the landscape will look like in three years. I'm being honest about that. But what I do know is that the architecture we are building is solid, the foundation is ready, the community driving it is fully committed and has the resources.

If you are evaluating whether Drupal is the right foundation for AI-powered content management, you don't have to figure that out alone. The Drupal AI Partners network brings together specialized agencies with deep experience deploying exactly these capabilities. If you are ready to move from evaluation to implementation, that is the right place to start.

We are all building in conditions none of us have navigated before.
The difference is what we are building on.

I've Seen This Movie Before.

Phase II Technology -

I've Seen This Movie Before. kdavis Thu, 04/16/2026 - 14:23

Back in October 1992, I walked into the Moscone Center in San Francisco for Interop, one of the biggest networking conferences of the era. I'd seen a lot of tech demos before, but one stopped me in my tracks.

A small booth was showing something called 10Base-T (Ethernet over ordinary Cat-5 telephone cable). Two computers, quietly exchanging data, surrounded by a deliberately hostile environment: electric motors running, vacuum cleaners humming, devices designed to generate every kind of electrical interference imaginable. And the data transfer was rock solid.

It doesn't sound like much today. But what I understood in that moment was that Ethernet had just escaped the lab. It no longer needed expensive, cumbersome thick coaxial cable snaked through walls by specialists. Now it could run over the same phone wiring already in every office building in the world. That demo didn't just show a better cable. It showed the beginning of the Internet revolution: the moment a transformative technology became something anyone could use, anywhere.

I hadn't felt that feeling again in thirty-plus years.

Until last week at HumanX 2026.
 

 

Welcome to the Next Revolution
HumanX brought together a remarkable cross-section of AI's most consequential voices: thousands of leaders, all under one roof at the same Moscone Center from Interop 34 years ago. The speaker roster read like a who's who of the AI world: Fei-Fei Li, Matt Garman, Bret Taylor, Andrew Ng, Ali Ghodsi, Vinod Khosla, Al Gore, Ray Kurzweil, and voices from Snowflake, Anthropic, NVIDIA, Perplexity, Zoom, Cursor, Salesforce, and dozens more. Speaker after speaker, I kept having the same déjà vu: agentic AI has been talked about for a while now, but this felt like the point where it stops being a topic and starts being a reality.

Not AI as a chatbot or a co-pilot that helps you write emails faster, but AI as an autonomous agent: something that can take a goal, connect to your systems and data, make decisions, take actions, and deliver results, all without a human clicking through each step.

The parallels to 1992 are striking. Just as 10Base-T democratized networking by making it accessible and practical, agentic AI is doing the same for intelligent automation. The question is no longer can AI do this? It's how do we actually deploy it at scale?

The Signal Through the Noise
Across every session and panel, a few themes surfaced over and over again:

  1. Agents are where the value actually lives. Every major platform company (Salesforce, AWS, Vercel, Sierra) agreed that generative AI's biggest ROI isn't in content creation. It's in agents that take action inside real business workflows: customer service, sales prep, code generation, HR processes. The companies winning are the ones moving from "demos" to production agents doing real work at scale.
  2. Adoption is a culture problem, not a technology problem. The companies successfully scaling AI aren't the ones with the best models; they're the ones that normalized AI use across every role, embedded it in existing workflows, and gave employees permission to experiment. The biggest adoption killers? Cultures where you have to be the smartest person in the room, or where using AI feels like admitting weakness.
  3. The workforce reskilling challenge is urgent and underestimated. Andrew Ng made a bold call: everyone should learn to code. Not because they'll write code by hand, but because AI makes it possible for non-engineers to build things, and those who can will dramatically outperform those who can't. Meanwhile, the gap between the pace of AI change and universities' ability to update their curricula is widening fast.
  4. Humans need to stay in the loop. Despite all the agentic enthusiasm, speaker after speaker was clear: humans still need to own the decisions that matter. AI handles the rote, the repetitive, the data-gathering, but accountability, judgment, and empathy remain human responsibilities. "People plus AI is a new way to work" was practically the conference motto.
     

Data Is the Hardest Problem, Still
If there was one problem that came up again and again, sometimes directly and sometimes quietly lurking under the surface, it was data. Specifically, the challenge of connecting AI to the right data, in the right context, with enough quality and trust to actually act on it.

The "garbage in, garbage out" problem is very real in the agentic era. When an agent can autonomously cancel an order, qualify a loan, or generate a report, bad data doesn't just produce a wrong answer, it can trigger a costly wrong action.

Several distinct data challenges kept surfacing:

  • Most enterprise data is unstructured and hard to use. The good stuff isn't in clean spreadsheets. It's in videos, PDFs, Slack threads, email chains, and Confluence pages. RAG search and vector databases help, but they don't fully solve the problem of extracting reliable, contextual intelligence from this kind of data at scale.
  • Data is siloed, and quality varies wildly. Agents need context from a dozen systems simultaneously, but those systems don't naturally talk to each other. And even when you can connect them, the quality of what's in them matters enormously.
  • Data governance adds another layer of complexity. Not everyone should see everything, and neither should every agent. Different people have different access levels across different systems, and when an AI tries to synthesize information across all of them, enforcing those boundaries while still delivering useful answers is a genuinely hard problem. It's not just a technical challenge; it's an organizational and legal one too.


I've been working through exactly these challenges firsthand. Our team has been building what we call an Intelligence Layer for client projects: an agent that connects to the full project ecosystem (Slack, Google Drive, Jira, GitHub, Salesforce, and more). The goal is to give anyone on a project team the ability to ask natural language questions about project status, technical decisions, client context, and get accurate, grounded answers.

It works remarkably well, until you hit the data relationship problem. When the same information exists in multiple systems, which source is authoritative? If Jira says a ticket is closed but the related GitHub PR is still open, what does the agent say? If a client question was answered in a Slack thread and later updated in a Google Doc, which is current? These aren't AI problems; they're data integrity problems that AI inherits.

Some vendors are tackling this directly. DevRev's "Computer" product, for example, is built around the concept of Computer Memory: a unified, AI-ready layer that ingests data from across your tools and systems (structured and unstructured) into a single source of truth that agents can query and act on. It's an approach I’m watching closely.

I’m Here For It
HumanX 2026 felt like Interop 1992: the moment a transformative technology became practical, accessible, and unstoppable. Agentic AI is no longer a research project or a vendor pitch. It's running in production at companies around the world, doing real work, at real scale.

Not everything you see today will survive. Not every agent platform, not every AI startup, not every use case will make it. That's fine. The dotcom era gave us a lot of Pets.coms, but it also gave us Amazon.

The question for every practitioner and leader isn't whether to engage with agentic AI. It's whether you're going to be in the driver's seat when the revolution arrives, or scrambling to catch up after it passes you.

I know which side I plan to be on.

Interested in talking more about the Intelligence Layer we're building, or how AI agents can be applied to your business?

I'd love to connect.

Publication Date Thu, 04/16/2026 - 14:04 Mike Potter Principal Engineer, CMS

Mike’s career started as an experimental neutrino particle physicist before creating the first WWW home page for Los Alamos National Laboratory. Mike has extensive experience architecting, designing, and overseeing the implementation of many complex enterprise solutions for Phase2, and also architected and led the development of the Open Atrium collaboration framework product.

Featured Blog Post? Yes Has this blog post been deprecated? No Summary HumanX brought together a remarkable cross-section of AI's most consequential voices: thousands of leaders, all under one roof at the same Moscone Center from Interop 34 years ago. Topic Artificial Intelligence HumanX Blog Banner Image Promo Image

The Drop Times: Erdfisch Expands nerdfisch DevBits into Public Drupal Code Archive

Drupal Planet -

Reusable fixes often remain confined to individual projects, forcing developers to solve the same problems repeatedly. erdfisch has expanded its internal DevBits system into a publicly accessible archive, exposing working Drupal code snippets drawn directly from project work. The collection prioritises immediate implementation over explanation, making internal solutions available without reshaping them into long-form documentation.

Drupal Starshot blog: Differentiating Marketplace Site Templates and Community Site Templates

Drupal Planet -

Site templates are available through two distinct pathways, each serving different needs within the community.

The official Drupal.org Marketplace provides a curated collection of site templates that meet certain quality standards, and are built on top of Drupal CMS as a foundation.

Community templates offer an alternative pathway for innovation and experimentation without the constraints of the curation process, by publishing the template as a general project on Drupal.org.

Official Marketplace Site Templates

The Drupal.org Marketplace are built on top of Drupal CMS, and curated to provide new users with confidence that they're starting with a consistent, solid and professionally built foundation that follows established best practices.

Key characteristics

  • Templates undergo a review processes

  • Must follow Drupal CMS best practices for security, accessibility (WCAG 2.2 AA), performance, and code quality

  • In the beginning, focus is solely on growing Drupal CMS adoption; site templates accelerate adoption of Drupal CMS by providing context relevant demo content and Drupal Canvas-compatible theme

  • Clear documentation, maintenance commitments, and user support expectations

  • Currently open to Drupal Certified Partners (for organizations) and Ripplemakers (for individuals or very small companies). Apply to become a creator here.

Benefits

  • Consistency for users who need reliable, production-ready starting points

  • Quality assurance through professional review processes

  • Support and maintenance commitments for long-term sustainability

  • Revenue opportunities for professional template creators

  • Sustainability for the Drupal Association through revenue sharing

Community Site Templates

Anyone interested in contributing a template can do so now, by publishing it as a general project on Drupal.org. All free site templates, including marketplace templates, are general projects for packaging and distribution purposes. Community site templates will be considered for inclusion in the Drupal.org Marketplace based on their compatibility with the outlined criteria.

Key characteristics

  • Can be published without formal review or approval

  • Not bound by the same standards as Marketplace templates

  • Can be built using Drupal CMS or Drupal Core

  • Available to all community members

  • Can take risks and explore directions that might not fit Marketplace criteria

Benefits:

  • Innovation by removing barriers to experimentation

  • Diversity of approaches and implementations

  • Learning opportunities for the community to explore what's possible

  • Stepping stones that might eventually evolve into Marketplace templates

  • Lower barriers to entry for community contribution

Security advisories: Drupal core - Moderately critical - Cross-site scripting - SA-CORE-2026-003

Drupal Planet -

Project: Drupal coreDate: 2026-April-15Security risk: Moderately critical 13 ∕ 25 AC:Basic/A:User/CI:Some/II:Some/E:Theoretical/TD:DefaultVulnerability: Cross-site scriptingAffected versions: >= 11.3.0 < 11.3.7CVE IDs: CVE-2026-6367Description: 

Drupal 11.3 comes with support for completing entity suggestions whilst adding a link to CKEditor 5.

The suggestions aren't sufficiently sanitized and a malicious user could trigger a stored cross site scripting attack against another user.

Solution: 

Install the latest version:

  • If you use Drupal 11.3.x, update to Drupal 11.3.7
  • Drupal versions below 11.3 are not affected by this vulnerability
Reported By: Fixed By: Coordinated By: 

Pages

Subscribe to www.hazelbecker.com aggregator