Blog
Writing on AI roleplay, manager conversations, and how teams generate scored, audit-ready proof of competence.
Pinned
Research
We tested the AI models underneath every roleplay platform on conversational realism, response speed, and feedback quality — then published the scores.
Read the benchmark→
16 models · August 2026
Thought leadership
Most trainers bolt AI roleplay onto old classroom courses. This guide shows L&D and consultants how practice-first design actually builds skill.
Thought leadership
A photorealistic avatar can still sound like a robot. Here is how we score every conversation — and why messy dialogue is what makes practice real.
Thought leadership
Scenario-based training defined plainly: four types, real examples, why it beats passive e-learning, and a seven-step build you can run this week.
Thought leadership
Organisations have spent decades measuring who attended training. AI role-play measures whether people can actually do the thing — and proves it.