1.1. Context: Complaining about Font Production Limitations

Type design occupies a distinctive niche within visual communication, where artistry meets technical precision in the creation of typefaces. The term “font” itself reveals an intriguing duality that bridges historical and contemporary practices. In traditional printing, a font comprises physical matrices – the tangible tools of typesetting. In the digital realm, it manifests as a software-readable document that enables the display of text in its designated typeface.

Visual communication design perpetually hungers for fresh modes of expression (Sowersby 2016; Burian, n.d.). While this creative appetite manifests through various media – photography, illustration, and other visual arts – typography stands as a particularly vital ingredient. Much like the cyclical demands for new films, music, or fashion, the typography industry responds to an endless quest for novel typefaces (Pflug 2020; “Global Font Use & Forecasting Survey 2024” 2024). These fonts, far from being mere technical tools, function as cultural artefacts that reflect both contemporary trends and individual creative voices.

This perpetual hunger for typographic novelty sustains an industry capable of delivering new fonts within a mere six months (Sonnad 2015) – complete with multiple weights, italic variants, and an arsenal of nine hundred glyphs spanning the Latin language system – a relatively modest scope compared to what follows.

While the Latin font designer can deliver new font in months, anything out of the Latin world struggles to ship font within a year (“How Long Does It Take on Average to Create a New Font?” n.d.; Sonnad 2015). Remember, nine hundred characters – the basic Chinese alphabet contains about five thousand characters. That means that the scope of the Chinese alphabet is larger in order of magnitude than Latin (Sonnad 2015). There is no doubt why most of the AI-based font generation attempts happened in China, Japan and South Korea – CJK 1.

Font production demands considerable expertise, effort and often collaborative teamwork – a complexity that limits our ability to type with the same expressive freedom found in handwriting. The constraints of typesetting and digital typing remain tethered to both the technical capacity of font production and the stylistic preferences of type designers. Just as programmers have embraced AI copilots (Bird et al. 2022; Barke, James, and Polikarpova 2023; GitHub Copilot · Your AI Pair Programmer” 2025) to augment their workflow, type designers might welcome an AI assistant capable of extending their master designs across an entire alphabet – a digital apprentice translating their artistic vision into complete typefaces – what a bright vision of the future.

Barke, Shraddha, Michael B. James, and Nadia Polikarpova. 2023. “Grounded Copilot: How Programmers Interact with Code-Generating Models.” Proceedings of the ACM on Programming Languages 7 (April):85–111. https://doi.org/10.1145/3586030.
Bird, Christian, Denae Ford, Thomas Zimmermann, Nicole Forsgren, Eirini Kalliamvakou, Travis Lowdermilk, and Idan Gazit. 2022. “Taking Flight with Copilot: Early Insights and Opportunities of AI-powered Pair-Programming Tools.” Queue 20 (6): 35–57. https://doi.org/10.1145/3582083.
Burian, Veronika. n.d. “Why Do We Need More Typefaces? | TypeTogether.” Accessed January 18, 2025. https://www.type-together.com/why-do-we-need-more-typefaces.
GitHub Copilot · Your AI Pair Programmer.” 2025. GitHub. 2025. https://github.com/features/copilot.
“Global Font Use & Forecasting Survey 2024.” 2024. Monotype. 2024. https://www.monotype.com/global-font-use-survey-2024.
“How Long Does It Take on Average to Create a New Font?” n.d. Quora. Accessed January 10, 2023. https://www.quora.com/How-long-does-it-take-on-average-to-create-a-new-font.
Pflug, Mary Catherine. 2020. “Results of the 2019 Font Purchasing Habits Survey.” Font Stuff. February 3, 2020. https://medium.com/font-stuff/results-of-the-2019-font-purchasing-habits-survey-39339f591a6c.
Sonnad, Nikhil. 2015. “The Long, Incredibly Tortuous, and Fascinating Process of Creating a Chinese Font.” Quartz. December 18, 2015. https://qz.com/522079/the-long-incredibly-tortuous-and-fascinating-process-of-creating-a-chinese-font/.
Sowersby, Kris. 2016. “Klim Type Foundry · Why We Need New Typefaces.” May 2, 2016. https://klim.co.nz/blog/why-we-need-new-typefaces/.

  1. In the chapter “Preliminaries”, it is clear that most of the font generation attempts come from China, following Japan, with a few attempts from Korea. Together, they overwhelm the Latin-based world with approximately 70/30 of all the published research papers.↩︎

Citation

If this work is useful for your research, please cite it as:

@phdthesis{paldia2025generative,
  title={Research and development of generative neural networks for type design},
  author={Paldia, Filip},
  year={2025},
  school={Academy of Fine Arts and Design in Bratislava},
  address={Bratislava, Slovakia},
  type={Doctoral thesis},
  url={https://lttrface.com/doctoral-thesis/},
  note={Department of Visual Communication, Studio Typo}
}