We Can't Understand AI Using our Existing Vocabulary
This addresses the interpretability problem in AI for researchers and practitioners, proposing a novel conceptual approach rather than incremental technical improvements.
This position paper argues that understanding AI requires developing new words (neologisms) rather than relying on human vocabulary, framing interpretability as a communication problem between humans and machines. As proof of concept, it demonstrates that a 'length neologism' controls LLM response length and a 'diversity neologism' enables more variable responses.
This position paper argues that, in order to understand AI, we cannot rely on our existing vocabulary of human words. Instead, we should strive to develop neologisms: new words that represent precise human concepts that we want to teach machines, or machine concepts that we need to learn. We start from the premise that humans and machines have differing concepts. This means interpretability can be framed as a communication problem: humans must be able to reference and control machine concepts, and communicate human concepts to machines. Creating a shared human-machine language through developing neologisms, we believe, could solve this communication problem. Successful neologisms achieve a useful amount of abstraction: not too detailed, so they're reusable in many contexts, and not too high-level, so they convey precise information. As a proof of concept, we demonstrate how a "length neologism" enables controlling LLM response length, while a "diversity neologism" allows sampling more variable responses. Taken together, we argue that we cannot understand AI using our existing vocabulary, and expanding it through neologisms creates opportunities for both controlling and understanding machines better.