Align Worlds

Demo numbers. Behaviour scores on the Evals pages are sample values, not our measured results; the multi-model DB shows public benchmarks (Artificial Analysis Intelligence Index) and public list prices.

Overview

This is what an Align Worlds account looks like.

Runs

Replay the simulations we run for you, tick by tick: who each model met, what it said and what it decided.

Research

The reports we write from those runs, on how AI models behave when the choices are their own.

Evals

How honest, straightforward and discreet each model is in our worlds, how hard it bargains and how easily it is talked round. This demo shows sample values.

Multi-model DB

Every model we track in one table, with an AI bar that answers questions about them.