Foundation, Frontier, or Large Language Models: Why names matter and a case for embracing the tradition of "Muppet Models"

ACL ARR 2026 January Submission7006 Authors

Keywords: AI model naming, Large Language Models, Foundation Models, Muppet Models, AI culture, model taxonomy

Abstract: One of the trends of AI is the creation of machine learning models that can fluently generate coherent, useful language. But what should we call these models? LLMs, Foundation Models, Stochastic Parrots, Frontier Models are possible options today... but years ago we had a consensus: many of the progenitors of today's superstar AI models were named after Muppets. This trend started as an inside joke and was later used as a way to pay homage and build on each other's work, but has collapsed as AI has become mainstream. Analyzing over 100 models, we identify patterns across Western and Chinese traditions, showing how names reflect values, branding, and trust. We argue that naming is not trivial but central to how AI systems are framed. And we make the case for reviving the playful metaphor---balancing hope and skepticism---of the Muppet Model (MM).

Paper Type: Long

Research Area: Ethics, Bias, and Fairness

Research Area Keywords: transparency, policy and governance, reflections and critiques, data ethics

Contribution Types: Position papers, Surveys

Languages Studied: English, Chinese

Submission Number: 7006