Home Platform Sectors Deployment About Contact Invest

Reliable, cost-effective automation for critical infrastructure

A new innovation in machine intelligence is bringing reliable, cost-effective automation to the critical infrastructure sectors.

These are the domains the technology was built for — mission-critical operations where a wrong answer costs an irreplaceable asset, a life, or the entire mission.

Request a demo Invest

16 sectors so vital their loss would be debilitating

The US Cybersecurity & Infrastructure Security Agency has identified 16 critical infrastructure sectors whose assets, systems, and networks, whether physical or virtual, are considered so vital to the United States that their incapacitation or destruction would have a debilitating effect on the physical security, economic security, national public health or safety, or any combination thereof.

Some of these sectors are actively under threat by external bad actors, others are at risk due to under capacity, deferred maintenance and rising costs. Cost-effective and rapidly deployable automation for these systems reduces operating costs, virtually eliminates human error and makes them internally robust to component failures and external threats.

01Chemical 02Commercial Facilities 03Communications 04Critical Manufacturing 05Dams 06Defense Industrial Base 07Emergency Services 08Energy 09Financial Services 10Food and Agriculture 11Government Facilities 12Healthcare and Public Health 13Information Technology 14Nuclear Reactors, Materials, and Waste 15Transportation Systems 16Water and Wastewater Systems

Human-in-the-loop judgment — with or without the human

Applications of our technology, called sapiens, can provide human-in-the-loop judgment and expertise — with or without the human.

Like highly trained human experts, sapiens make decisions that can be trusted, explained and verified. Unlike human experts, sapiens never miss a data point, never get distracted, and never forget — anything.

How it works →
Plant or mission operations, wide, human in frame.
Operations photography Plant or mission operations, wide, human in frame. img/feature/operations.jpg
0

CISA critical infrastructure sectors a sapiens can serve

0+

Patents held collectively across the leadership team and board

0

Data centers required. A sapiens runs on a mobile device processor

This is not research-stage technology

The approach has been delivered in production before, on earlier generations of the same architecture, for NASA, Boeing, and the U.S. Navy. Real operational use, not simulations.

Hubble-era mission control, operators at telemetry consoles.
Mission operations Hubble-era mission control, operators at telemetry consoles. img/proof/hubble-space-telescope.jpg
NASA / Goddard

Hubble Space Telescope

45 → 2 minto interpret spacecraft telemetry, down from 45 minutes

The approach originated with founder Bryant Cruse’s work on the Hubble Space Telescope at NASA/Goddard, where the first AI system for spacecraft telemetry analysis reduced interpretation time from 45 minutes to under 2 minutes.

Delivered on prior-generation architecture
Launch vehicle on the pad, or launch control room during a countdown.
Launch pad Launch vehicle on the pad, or launch control room during a countdown. img/proof/conestoga-launch-vehicle.jpg
Launch control

Conestoga Launch Vehicle

2 catcheserrors found that human operators missed

The system twice detected errors overlooked by human operators. It automatically held a countdown after detecting an explosive hazard operators had missed. On a later attempt it correctly identified an anomaly as a non-critical sensor failure, clearing the vehicle to launch on schedule and saving days of delay and over $1M in recycling costs.

Delivered on prior-generation architecture
Satellite ground station or constellation operations floor.
Satellite operations Satellite ground station or constellation operations floor. img/proof/uhf-follow-on-constellation.jpg
U.S. Navy

UHF Follow-On Constellation

1 laptopbeside a room of IBM mainframes

Deployed for the U.S. Navy’s UHF Follow-On satellite constellation, it ran on a single laptop beside a room of IBM mainframes. Air Force personnel repeatedly came over to watch its displays because they made spacecraft status so much easier to read.

Delivered on prior-generation architecture

This isn’t machine learning. This isn’t symbolic AI.

It is synthetic intelligence — and the difference is not a matter of degree.

This isn’t

Machine Learning

Machine Learning uses statistical patterns to simulate intelligence the way a video of fire simulates fire: it’s a realistic likeness, but the video is not on fire. What’s missing is precisely what’s essential.

If you want rationally competent machines, you don’t need a bigger, better camera; you need a different kind of instrument.

Sapiens operationalize human understanding of how the world works without a natural brain, just as an airplane’s wings replicate the abstract structure of aerodynamics without any bird.
This isn’t

Symbolic AI

Symbolic AI applications like expert systems or semantic network ontologies are based on collections of rules with no principled account of how knowledge is structured. This is why they are brittle and difficult to scale beyond small applications: a heap of rules is not a model of a domain.

A sapiens holds a world model grounded in a robust epistemology; it’s a dynamic map of how the domain is and behaves rather than a list of heuristics that only work inside it, under static conditions.
It is

Synthetic Intelligence

New Sapience has developed a platform for scalable, deterministic, world-model-based applications in which knowledge is achieved as non-symbolic, directed graph structures.

Our IDE enables commonsense and expert knowledge derived from the customer’s domain and requirements to be reconstructed in computer memory as synthetic knowledge.

Synthetic Intelligence is real knowledge but not contained in a natural mind — in the same way a synthetic diamond is a real diamond but not created by natural processes.

Latest news

Under the Hood: Sapiens and Large Language Models ComparedPaper All news→

Tell us about your system that cannot afford to be wrong

Request a demo