Google has launched Vantage, a research experiment that leverages generative AI to assess future-ready skills through simulated conversations. Developed with New York University researchers, the tool creates dynamic environments where high school and college students demonstrate competencies like critical thinking and collaboration that traditional tests often miss.

The system uses AI avatars in open-ended scenarios, with an Executive LLM steering conversations to target specific skills. An AI Evaluator then analyzes transcripts against pedagogical rubrics. Validation studies with 188 testers showed the AI's assessment accuracy comparable to human experts.

Vantage addresses the challenge of measuring skills that education systems struggle to quantify. The research explores transforming these "hard-to-measure" competencies into assessable abilities at scale, potentially creating a measurable "skills layer" for school curricula that develops both academic knowledge and durable human competencies.