Monthly Community Update - January 2026

Announcements
1

Hi everyone,

Happy New Year! This is the first community update of 2026! You can read December’s update here.

Where we’re headed in 2026

2026 is the year we prove to the world that AI makes no-code visual development more, not less, important.

Successful builders understand what they’re creating. You can vibe out a prototype without knowing what’s under the hood, but if you want to build a scalable, secure, fully-featured app — and turn it into a profitable business — you need to be able to understand, debug, and evolve it, even when the AI gets confused or hits a dead end.

Bubble makes it easier, faster, and simpler to build robust apps. That said, that initial “wow” moment of vibe coding an app is real, and visual development platforms ignore it at their peril.

This year we’re going all in on our AI Agent to make the Bubble “wow” moment pop, too (no pun intended). The difference being our AI will increasingly teach users how to understand and edit what’s being built. Investing in AI is critical to the future of the platform and the best thing we can do for our long-term success of the community. But we also understand that for experienced Bubblers who already get why Bubble is special, building faster with AI isn’t your top wishlist item for us.

So rest assured, we’re also going to be spending time in 2026 working on things that matter to our power users and scaling businesses. That includes continuing to fix usability pain points with the editor, continuing to mature our mobile offering, making it easier to build LLM-powered apps on Bubble, and continued investment in our platform’s reliability and scalability. We’re in the midst of building out an entire subteam of engineers focused on the developer experience. All of this work means that everyone, from long-time users to those just getting started, can feel confident that they can use Bubble to turn their app idea into a profitable business.

We have a lot of building to do in 2026, and I’m excited to get to it. We appreciate all the support! Now onto the updates.

What we shipped in December

We upgraded the Agent’s AI model to OpenAI’s GPT 5.1, resulting in 20–30% faster response times as well as improved component generation and editing. Bubble uses different models in different situations in order to produce the best results. For example, we’re exploring GPT 5.2 but our testing hasn’t shown performance gains that make this update feel worthwhile yet. We’ve also upgraded app generation to Anthropic’s Opus 4.5, which has improved overall output quality.

We completed work on a behind-the-scenes refactor to resolve duplicative Agent responses. This work also sets the foundation for compound edits (asking the Agent to do two things in the same request). Additionally, we finished work to migrate the Agent’s knowledge base which allows us to improve response times as more people use it.

All users now have access to mobile app generation! Bubble AI can create mobile front-ends, including dynamic expressions that help lay the groundwork for building out the app’s backend. We also made some updates to the prompt box for app generation so that you can select whether you want to generate a web or mobile app. You can read more on the forum here.

In November, we released an update that allows users to toggle between mobile and web workload usage in your app metrics tab. As a follow-on, we shipped an update that lets you toggle between web and mobile server logs as well.

The beta version of the redesigned property editor went live for 25% of existing and new users. We’re working on feedback from this first batch before rolling it out to more users. We also launched a comprehensive migration guide to help power users get familiar with the new interface. The team also recently shared an update from the first phase of the beta, which includes a roadmap for planned fixes based on your feedback. You can read it here.

We added two new entry points to the security dashboard: one in the editor sidebar and one when you’re getting ready to deploy your app. The security dash scans for data leaks, exposed API keys, or unsafe API configurations, and other vulnerabilities. Read more on the forum here.

We wrapped our migration of the editor’s source code to Typescript, which makes the editor easier and safer for us to update.

We made some updates to the UX for scheduling workflows. Runaway workflows (where the user accidentally schedules thousands of the same workflow) is one of the biggest causes of emergency performance and WU consumption for our users. In the past, this has generally resulted in outreach to our team for support. The new interface lets you address this yourself: A notification on the scheduled workflow ta b will show your most-scheduled workflow and let you cancel all instances of it without impacting your other workflows.

scheduler
scheduler1600×256 50.5 KB

We redesigned the Academy to make it easier to see important course details like level and number of lessons. We also launched interactive tutorials, which walk you through the lesson step-by-step. It also looks a lot more modern, too! Phase 2 of Academy work, course progression tracking, is coming this month.

The new experts hub is live as well. For users looking for help, either via coaching sessions or hiring an agency/freelancer, it’s easier to find the right help. For coaches, there’s now a coach profile with a unique URL so it’s easier to link and share your work. You also have more payment flexibility so you can link your preferred service. You can read more about both of those updates here.

For a full list of bug fixes and improvements, you can check out our Releases page here.

What we’re working on in January

  • AI Agent: The team is working on bringing data type and workflow generation to the AI Agent later this month. Soon the Agent will be able to create, edit, and reference data types and create frontend workflows. That means the agent will be truly full-stack and have the ability to help you throughout essentially the entire build process.
  • In-app purchases: This will be available for broad beta testing starting in February. In-app purchase functionality is a critical unlock for mobile developers who want to monetize their apps and it will work for both main cluster and dedicated apps at launch. If you missed it, you can view the demo here.
  • API Connector: We’re making the API Connector easier to find later this month, rather than burying it in the plugins tab. This is especially important as more users connect to AI models to build functionality.
  • API workflow permissions: We’re adding an option for backend API workflows to be runnable only by users with admin API keys. Users will be able to select if API workflows can run without authentication, only by authenticated users, or only by admins. This update should be live in the next week or so.
  • Security dashboard: We are sunsetting security monitoring features at the end of January. This is a feature with a very small number of users that has an outsized load on our system. If you are an active or previous user of this feature, you should have received a reminder email with recommendations for alternate tools. You can read more about this update here.

We’re also working on how raised security issues are displayed in the dashboard. The current configuration can be overwhelming and bury more important issues. The revamp will re-categorize results from security checks by severity and issue type to make them easier to resolve. That update is coming later this month.

How to get involved

  • RSVP to the next founder AMA with Emmanuel on January 15.
  • We’re looking for more customers to highlight, including via our CEO Chat series with Emmanuel. Fill out this form if you’re interested: bble.io/story.
  • Cleber is Mexico’s largest auto dealer group and they built several solutions on Bubble, including a customer-facing marketplace, in a super short timeline. Read their story here.
  • Check out the interactive map of top apps from around the world! This is a really cool initiative that showcases the community and just how many awesome apps are built on Bubble
  • If you’re a regular reader of this update, you probably also received personalized Bubble Wrapped stats. Make sure to check your email (and share on social!). The global stats were impressive, too. Your apps ran a combined 28.6 billion workflows in 2025. Read the full wrap up on the blog.
  • There’s a community meetup happening in Arkansas on January 21. RSVP here.

New hires

We had two new hires join us to close out the year. Welcome to Aidan (software engineer) and Emmelie (growth marketing). We’re hiring for a number of roles right now. Check them out on our Careers page!

That’s all for this month.

— Josh and Emmanuel

17 Likes
Timeline update for native mobile?
3

Can’t wait to see what you’re about to do with the API Connector! It’s about time :tada:

1 Like
4

Josh, I get the ambition here, and I agree that visual development matters more, not less, in an today’s AI world.

Where this starts to feel disconnected from reality is the idea that the Agent is close to being “truly full-stack.” Right now, it doesn’t really produce anything you can take and use in a real Bubble app. Other than a rough, non-modular UI scaffold that most people throw away, there isn’t much output that survives once real building starts.

That’s why the full-stack framing feels premature. The issue isn’t autonomy, it’s utility. The Agent doesn’t meaningfully help with the work people actually do every day in Bubble.

If this is going to matter, it feels more impactful to treat the Agent as a set of practical editor tools. For example, generating tables / repeating groups that actually work: real data, real columns, sorting, filtering, and basic formatting. That’s the stuff most apps are made of.

Even setting output aside, the current ~2K character prompt limit makes the full-stack aspirations hard to square with reality. You run out of room long before you can describe anything close to a real app, let alone iterate on it in a meaningful way.

These everyday tasks are unglamorous, but they’re most of the work. They’re also where time gets burned and mistakes happen.

Right now it feels like we’re skipping this layer and jumping straight to full-stack AI builder. If the Agent got genuinely good at these fundamentals first, it would actually change how people use Bubble. Everything more ambitious would have a much stronger foundation after that.

11 Likes
5

You know what would be both useful and convenient? Just like we have a Monthly Community Update, where updates are shared with the entire platform, it would be very valuable to have something similar focused on Native Mobile progress. Maybe a Monthly Community Mobile Update? This would help keep everyone informed about advancements and priorities in the mobile space.

6 Likes
6

You mean like this?
The codeless extension already does exactly this, which kind of highlights how overdue and low-effort the change actually is.

image
image107×436 4.21 KB

1 Like
7

Awesome QoL updates :raising_hands:

4 Likes
8

A built-in shortcut wouldn’t hurt, but I expect we’re talking about something more substantial. Otherwise it’s not even worth a mention in the monthly update.

2 Likes
9

My take is that,

  1. GPT is the inferior one of those 3, and they are consistent with it.

  2. I recently removed all my plugins and replaced it with code, using AI, I always had a part where I need to tell it to use static data, due if you make any mistake I wont be able to replace it because text editor doesn’t allow dynamic values to be copy and pasted, so I wasted too much time on making them static then converting to dynamic, then again solving some errors due to delusion.

So allowing dynamic values to be pasted if the syntax is correct would be nice if it is really easy for you to push new changes.

  1. I think native Bubble elements does not suffice the AI roadmap, in parallel you need to update them.
10

I’ve been seeing people asking about branches in workflows for a while now, and loops would also be very welcome. There are more complex features where it would be wonderful to have these options; they are basic programming logic, but complicated to implement in Bubble.

I get anxious every time a new month arrives, hoping they’ll announce this, but it never happens.

6 Likes
11

An old promise that was not fulfilled and also has no answer.

12

An old promise that was not fulfilled and also has no answer.

2 Likes
13

The three features that would be truly game changing:

  • Ability to choose async/sync workflows (and ideally functions) - this is by far the most infuriating aspect to bubble is trying to work out under what conditions things run, as it is dynamic. At the very least, TELL me how they will run to save me the hassle!
  • if/else conditional branching
  • Loops

All of the above need to exist both client side and server side.

I personally find zero value in AI for anything other than support chatbots that can help by searching forums and documentation. In every other use case I spend more time validating and correcting the numerous issues created by flaws in the design or logic.

I have yet to get anywhere with mobile apps due to the lack of equivalent plugins, very frustrating, but hopefully that’s on the way…

8 Likes
14

Im still antsy for the new editor, flaws and all. Disappointed we can’t voluntarily opt-in if we are outside of the 25% group. I have several Bubble apps I’m working on, and none were lucky enough to be in the test group. I will not be buying a lottery ticket anytime soon.

The API changes sound promising. I am currently building out APIs to connect to Zoho and a few other platforms, so the ability to be more granular about permissions is a welcome update.

15

Bubble’s AI Agent being restricted to apps created through the AI generator is a major limitation.

If Bubble is positioning AI as the future of the platform, it feels counter-intuitive that the benefits are only accessible to users starting from scratch. The reality is that thousands of existing production apps could be improved dramatically if AI could also accelerate component building things like:

  • tables and data layouts

  • repeating groups

  • structured dashboards

  • bulk UI scaffolding

Right now, the core value proposition seems to be: “Describe an app, generate it, and then AI can help you edit it.” That’s useful for prototyping, but it doesn’t serve teams maintaining or scaling real apps, or founders who want AI to speed up development of specific sections rather than regenerate their entire product.

The opportunity here is obvious: AI that enhances what already exists, not just what gets generated. That would be a true platform unlock, instead of a segmented advantage.

On top of this, continued downtimes in recent months raises questions around reliability for production workloads. AI innovation is exciting, but without production-grade stability and broader accessibility, it risks becoming a feature that benefits a small subset of new users, while the majority see little to no impact.

Would love to see Bubble expand AI to:

  1. work with pre-existing apps, and

  2. accelerate building individual UI/data components, not just full app generation.

That’s where the real value is for the ecosystem.

2 Likes
16

Thanks @josh for the update, these are some nice changes. It’s always good to hear your thought process on the roadmap and how Bubble should move forward. Bubble is one of the few companies where it feels like they want to have a dialogue with their users.

This should be the most important objective of 2026. As I’ve said before, it’s been a full decade since 2016, during which time every niche that can be conquered has already been conquered. LLM-powered apps are the final frontier. I use the example of the “laundromat point of sales software” of @BrianHenderson fame. If he had tried to create that today, it would be unsuccessful because he would be competing with his already established self. This is not necessarily a Bubble problem but an industry problem. It’s why even giants like OpenAI and Google are scrambling to build “AI automation” into their core products. Workflows, etc. The amount of business Bubble has lost to n8n since 2023 could have been completely avoided with the right focus (making Bubble JSON-native, allowing for parallelization of LLM calls, etc.).

@emmanuel recently stated that AI-generated Bubble apps have less churn than regularly-created Bubble apps, presumably for this reason. In my opinion you are focusing on the wrong type of churn. I would like to distinguish “effective churn” from “raw churn.” Raw churn contains users who would have stopped building on Bubble even if the product was perfect, simply because they don’t actually have the motivation/work ethic/talent to build a company from scratch. These are just tinkerers and opportunists looking for the next “easy buck.” Effective churn is what you need to look at. The best place to see this is the conversion rate from Starter –> Growth and Growth –> Team. These are your highly profitable pricing bundles that make up for the Free/Starter “loss leaders.”

I suspect that these numbers have gone down from their historic highs. I obviously don’t have the internal data to prove it, this is just based on foresight. The mechanism for this, I believe, is two-fold:

  1. Users start building LLM-powered apps and hit major roadblocks with a lack of parallelization and poor/unintuitive JSON handling. They decide it’s too much work to build a proper AI-native app in Bubble, pack their stuff up, and go somewhere else. This is the most common reason. It’s ironic because it’s analogous to vibe coding an app and hitting a roadblock, which is exactly what Bubble is trying to “fix.”

  2. Users hit product market fit, costs go brrrrr, they decide it’s time to go with something cheaper rather than more nimble (since the features have already been built). I think this is less common than #1 and applies more to the Enterprise tier than anything else (which is why we need a tier between Team and Enterprise.). The best way to fix this is to prevent users from tinkering with plugins and code. It’s a slippery slope. Any time you require users to familiarize themselves with Javascript or 3rd party services (Cloudflare Workers, etc.), they get a little bit more comfortable moving off the app. You need to adopt the Apple mentality of, “every single element is fully-featured and polished” and stop thinking about the plugin marketplace as a monetization opportunity. This is why Apple slowly integrated all the major apps (i.e. plugins) into the core of iOS over the course of 10-20 years.

Anyway, focus on fixing #1 rather than trying to capture “inevitable churners.” I also think part of the problem is wrong targeting. You should be aiming for vibe coders who hit roadblocks/security issues and focus the marketing on that. Kind of like the TV show “Botched.” You want to position yourself as the way to fix their useless apps, not as “just another” product that creates useless apps.

Lastly, and this is addressing the 100,000 pound herd of elephants in the room: I think the entire AI agent is misguided and not even technically feasible. Bubble is legos. You’re trying to combine “3d-printed” code with legos that have very specific tolerances and ways of fitting together. It’s never going to be safe to edit an entire application. This is ignoring the fact that no context window is going to large enough for some of the larger apps’ JSON files.

Screenshot 2026-01-06 184549
Screenshot 2026-01-06 1845491435×1101 60.9 KB

The AI editor needs to be an “airlock.” You click on an Action, Thing, or Element and say “Edit in Assistant.” This sends it over to a sandboxed environment where now the chat-based assistant can edit it freely without worrying about breaking the entire app. You can’t let the LLM decide where to place the code snippets in the JSON. You also can’t let the LLM have access to the entire JSON. Instead you create specific element templates with strict guardrails where the chat assistant inserts dynamic data into the static template. That way you can safely create whatever you need in the validated environment of the sandboxed assistant, and then copy/paste it directly. The user decides where it goes and if it’s good enough. This also has the nice effect of not depending on the buggy undo/redo feature to roll back multiple changes.

Please understand that I took the time to write this post for free because I care about the future of your company as much as you do.

9 Likes
A bittersweet goodbye: Why I’m moving on after 5 incredible years on Bubble
17

Hi code-escappe,

Cory here, I’m a PM on the AI team, and I really appreciate this feedback! You’re definitely right that the Agent is still in beta and needs further improvements before it’s truly “full stack”. However, I do think you’re mixing up our ‘Agent’ with the ‘AI App Gen’ feature where you enter a prompt and get an app. Have you used our Agent that’s in the editor? It’s only available for AI apps as it’s still in beta, but it can create UI (repeating groups just like your request), expressions and in the coming days workflows and data schema as well. I’d love to pick your brain and get more feedback when you have a chance to use it more!

1 Like
18

Couldnt agree with you more! I‘m a PM on the AI team and we’re working hard to remove the limitation of the agent to AI apps only and in doing so it’ll be able to do exactly what you requested – build components but also workflows, expressions, and data schema.

19

What’s the status regarding IAP? We are releasing our app this Monday. It was finally accepted by Apple after a lot of back and forth, but the result is that we have to release the app totally free for our users. This means we are running a business while losing money every single day. Any updates?

20

@mikke2 here: