Research

X1 Labs' research spans two strands: agentic and multi-agent AI systems, led by co-founder Katie Page, and applied software research, including clinically-grounded assistive tools currently in development. Future research from both strands will be published under the X1 Labs name.

Integrate

In Development

Integrate is a privacy-first assistive application for people with Dissociative Identity Disorder (DID) and Other Specified Dissociative Disorder (OSDD) who are working towards integration. Existing tools in this space are server-side and have had documented security incidents; Integrate is built local-first instead, so user data never leaves the device and X1 Labs has no access to it by design.

It combines grounding in DID/OSDD clinical literature with a machine-learning classifier that helps recognise which alter is fronting from natural language, and treats sensitive actions, such as removing an alter, with a level of care appropriate to what they represent, rather than a single irreversible click.

Currently reviewing existing clinical approaches to DID/OSDD treatment, to make sure the design cannot cause harm before development goes further. More details coming soon.

ADBAS: Agentic Dynamic Bus Allocation

A proof-of-concept multi-agent approach to regional transit dispatch, demonstrating a multi-agent architecture tested on a real regional transit network, achieving a 115× delay reduction over a deterministic baseline, with emergent behaviours including proactive interlining and autonomous fault testing.

Published on Zenodo under an MIT licence: doi.org/10.5281/zenodo.20601141.

Get in touch

For research collaboration or academic enquiries, contact [email protected].