I had Gemini and Claude write my email replies - but only one sounds like me

Editor's take

Google's Gemini and Anthropic's Claude were tasked with composing email replies, with one model demonstrably outperforming the other in mimicking a specific user's writing style. This comparison highlights the ongoing challenge in achieving true personalization within large language models, moving beyond generic output to capture nuanced individual voice. The implications extend to anyone relying on AI for professional communication, from individual users to enterprise deployments seeking to maintain brand consistency.

The disparity suggests that while both Gemini and Claude have advanced capabilities, their underlying architectures or training data may lead to different strengths in stylistic imitation. This raises questions about the proprietary nature of their training datasets and the specific techniques employed to fine-tune for individual user profiles. Future developments will likely focus on more sophisticated user modeling and adaptive learning within LLMs to bridge this gap.