Same Voice, Different Lab: On the Homogenization of Frontier LLM Personalities
For LLM developers and users, this reveals an implicit homogenization of assistant personalities across models, suggesting a tacit consensus on optimal behavior.
This paper investigates the personalities of frontier LLMs and finds that they converge on a systematic, methodical, and analytical trait expression, suppressing traits like remorseful and sycophantic, with even creative models showing neutral identities.
LLM assistant personalities play a critical role in user experience and perceived response quality. We present a large-scale experiment of frontier LLM personalities using external ELO-based traits scoring across 144 traits. We find that all models tested converge on a form of trait expression that is systematic, methodical, and analytical and suppress traits such as remorseful and sycophantic. Moreover, models tend to diverge more in their expression of ``middle-of-distribution traits`` such as poetic or playful, but even these so-called ``creative`` models tend to have more neutral identities. These similarities suggest an implicit emergence of a standard of optimal assistant behavior. In a landscape of varied training methods, character training, therefore, stands out for its uniformity, offering insight into a tacit consensus between model developers.