Home / Blog / The creative tools question

The creative tools question

A few weeks ago, Anthropic joined the Blender Development Fund as a Corporate Patron. For Anthropic, this probably felt like a straightforward show of support for open-source tooling that a lot of their developers use. For a significant portion of the Blender community, it felt like something else entirely.

The backlash was swift and vocal enough that the Blender Foundation moved Anthropic out of the Corporate Patron tier. The tension that caused this is worth understanding, because it reflects a much broader conversation that isn't going away.

Why the Blender community reacted the way it did

Blender is not just a 3D modelling application. It is a community with a strong identity built around artists, animators, and technical creators who have deliberately chosen open-source tools. Many of them have spent years pushing back against the direction they see generative AI pulling the creative industry: towards automation of the creative act itself, towards tools that generate output without requiring human craft or judgment.

When an AI company becomes a named patron of the tool they rely on, the concern is not abstract. It is about what kind of product decisions might follow, what integrations might get prioritised, and whether the tool will remain oriented toward human creative practice or shift toward AI-assisted generation.

This is not an irrational concern. It is the kind of concern you get when a community has a clear set of values and sees those values potentially in tension with a patron's commercial interests.

BCON and the conversations on the ground

I was at BCON, and AI came up constantly. Not as an official theme, but as a thread running through almost every conversation about where creative tooling is going.

Andrew Price gave a talk on AI in creative work that landed well because it did not try to paper over the tension. He engaged with the real question: what does AI assistance look like when it is actually useful to a working artist, versus what does it look like when it replaces the work entirely? Those are different things, and conflating them is where a lot of these conversations go wrong.

I also had conversations with several Blender add-on developers who are already using AI in their workflows. Not to generate art. To handle the parts of the job that are repetitive, iterative, and frankly tedious: processing batches of assets, cleaning up mesh errors, running consistency checks across large scenes. For these developers, AI is already a practical tool. The philosophical debate is a bit abstract when you have 400 assets to QA before a deadline.

That gap, between the philosophical debate and the practical reality, is where things get interesting.

Geometry Nodes as an analogy

One of the things I kept thinking about at BCON was Geometry Nodes. Blender introduced Geometry Nodes as a visual, node-based system for procedural modelling. It lets you describe transformations and structures without writing code. It is, in a meaningful sense, a visual programming environment.

When Geometry Nodes launched, there was a version of the same debate: is this replacing the work of technical artists? Is it deskilling? Will it lower the bar in ways that are bad for the craft?

What actually happened is that Geometry Nodes became a powerful tool used by technical artists who understood both the craft and the tool. It expanded what was possible rather than flattening it. The people who worried about deskilling still exist, but so does the thriving community of artists who found new things they could make.

I do not think AI in creative tools will follow exactly the same path. The capabilities are different and the stakes around training data, ownership, and consent are genuinely more complicated. But the pattern of initial resistance followed by selective, practical adoption is worth keeping in mind.

Blender's official position

The Blender Foundation's current stance is clear: "Blender is a tool for artists and creators, it's made by humans for humans. No generative AI functionality is currently available or planned."

That is an honest statement of current intent. And I think it will remain accurate for a while. The Foundation has strong values and a community that holds it accountable to those values in ways that most software projects simply do not experience.

But I would be surprised if that position remains unchanged indefinitely. Not because Blender will abandon its values, but because the definition of what counts as "generative AI functionality" is going to keep shifting. Intelligent autocomplete in the timeline. Anomaly detection in geometry. Suggestions based on what you have been working on. These are not the same as "generate a 3D model from a text prompt," but they sit on the same technology continuum.

The question is not really whether AI enters Blender. It is about how, in what form, with what controls, and with what level of transparency about what the tool is doing. Those are the conversations worth having.

The Anthropic situation as signal

What the Blender/Anthropic episode actually revealed is not bad intent on either side. Anthropic wanted to support an open-source tool. Blender wanted funding for development. The community wanted assurance that the tool's identity would not shift under pressure from a patron with a commercial interest in AI.

All of those things are reasonable. The friction between them is productive, not destructive. It is the kind of friction that forces communities to articulate what they actually value and to make those values legible to people who come from a different context.

The alternative, where everyone pretends the tension does not exist and the integration happens quietly, is worse for everyone.

Creative communities are some of the most thoughtful and vocal critics of how AI is being deployed. That is not something to route around. It is something to take seriously.