Placeholder copy · the text on this page is a draft

Research

White papers on the problems industrial AI has to solve before it is worth trusting, and case studies from the sites where we tested the answers. Written by the engineers doing the work.

Featured white paper

Every industrial revolution since 1788 has handed one more control loop to a machine. This paper traces the one that never moved, and sets out what a system would need before a plant could trust it with that call.

Name SurnameCo-founder, Ferry
Featured white paper

Every industrial revolution since 1788 has handed one more control loop to a machine. This paper traces the one that never moved, and sets out what a system would need before a plant could trust it with that call.

Name SurnameCo-founder, Ferry

Latest

All

All

White papers

White papers

Case studies

Case studies

Field notes

Field notes

Alarm rationalization is a data problem, not a discipline problem

For four years Reactor 3 raised a high-pressure alarm that never once corresponded to a high pressure. Two engineers had tried to trace it and stopped. Time to root cause: forty minutes.

Read the article

Alarm rationalization is a data problem, not a discipline problem

For four years Reactor 3 raised a high-pressure alarm that never once corresponded to a high pressure. Two engineers had tried to trace it and stopped. Time to root cause: forty minutes.

Read the article

The historian already knows what happened last Tuesday

Shift handover loses information every time it is written from memory. The record to reconstruct it from has been sitting there all along.

Read the article

The historian already knows what happened last Tuesday

Shift handover loses information every time it is written from memory. The record to reconstruct it from has been sitting there all along.

Read the article

Agents and control systems

What has to be true before software is allowed to act on a plant, and why we have not let Mainframe do it yet.

Read the paper

Agents and control systems

What has to be true before software is allowed to act on a plant, and why we have not let Mainframe do it yet.

Read the paper

Sovereign deployment for industrial AI

The questions an OT security team should ask before any model reads plant data, and the answers we give.

Read the paper

Sovereign deployment for industrial AI

The questions an OT security team should ask before any model reads plant data, and the answers we give.

Read the paper

New research, once a month.

One email when a white paper or case study is published. Nothing else, and the list is never shared.