Aligned: A Platform-based Process for Alignment
This addresses the need for a more trustworthy, public-facing approach to AI safety, though it appears incremental compared to existing closed-door efforts.
The paper tackles the problem of aligning frontier models and superintelligence by introducing Aligned, a platform for global governance and alignment, resulting in a 30-guideline constitution with 93% overall support from 680 participants.
We are introducing Aligned, a platform for global governance and alignment of frontier models, and eventually superintelligence. While previous efforts at the major AI labs have attempted to gather inputs for alignment, these are often conducted behind closed doors. We aim to set the foundation for a more trustworthy, public-facing approach to safety: a constitutional committee framework. Initial tests with 680 participants result in a 30-guideline constitution with 93% overall support. We show the platform naturally scales, instilling confidence and enjoyment from the community. We invite other AI labs and teams to plug and play into the Aligned ecosystem.