Thoughts on Creative Automation

I have been thinking a lot about where the line is between using a tool and being used by one. This is not a philosophical exercise -- it is a practical question that affects how I work every day.

The Automation Spectrum

Not all automation is equal. There is a spectrum that runs from mechanical repetition (batch rendering, file conversion) through intelligent assistance (auto-keying, smart selections) to generative creation (AI-generated imagery, procedural storytelling). The first category is unambiguously good. The last is where it gets complicated.

Automation that removes drudgery is liberation. Automation that removes decisions is something else entirely.

The danger is not that AI will make bad art. The danger is that it will make acceptable art so easily that we stop pushing for great art. When "good enough" is free, the economic incentive to pursue excellence weakens.

Finding the Balance

My personal rule is simple: automate the execution, never the intention. If I can clearly articulate what I want and why, then using AI to get there faster is just efficiency. But if I am using AI because I do not know what I want -- hoping it will show me something interesting -- that is a different relationship.

My automation decision framework:

  [Task] → Is the output predictable?
    ├── YES → Automate fully (batch ops, formatting, exports)
    └── NO → Does it require taste/judgment?
              ├── NO → Automate with review step
              └── YES → Keep human, use AI for options only

The framework is not about rejecting technology. It is about maintaining agency. The best work comes from artists who use every tool available but remain the author of every decision that matters.