SEM.ai
SEM.ai Labs

AI Experiments

We run hands-on experiments on how Claude AI and LLMs actually behave. Between end users and researchers — reproducible setups, real data, practical findings.

#001Claude Behavior Series

Why Claude Gives Different Answers to the Same Prompt — A Session Independence Experiment

We ran the same prompt across 5 independent Claude sessions with identical model and system prompt settings. Here's exactly where the outputs converged — and where they diverged.

Claude AILLM non-determinismsession independence