hexagons containing images of insects, microbes, bat, flower

Microsoft Premonition

Biological monitoring and intelligence

Contact us for availability

Available to select partners as part of Early Access Program

Preventing diseases, maintaining food security, and protecting ecosystems require new tools to monitor the millions of hidden species in the environment.

Key: Pathogen risks Ecosystem services Food security risks

Microsoft Premonition is a state-of-the-art platform to identify, track, sample, and forecast key species and biological threats.


Platforms and capabilities

close up image of a finger pointing at a screen of metagenomics data

Digital and genomic species identification AI

We provide industry’s most comprehensive metagenomics AI for identifying species using genomic signatures — detecting everything from viruses to vertebrates. We also identify invertebrate species, which make up most animal biodiversity, in real-time using neural net powered digital sensing platforms. Our AI systems are deployed into harsh environments and applied to large and complex genomic datasets.

photo of a biological weather station in a forested area

Biological weather stations

Our biological weather station capability is a fully programmable and extensible bio-sensing and bio-sampling solution for cost-effectively monitoring critical and remote environments. It draws in major disease vectors and other species with high economic impacts, then identifies them in real time, and transmits high-resolution data back to the cloud. Uniquely, its advanced robotics capabilities allows it to collect targeted species for later molecular and genomic analyses, while leaving non-target species unharmed.

close up photo of a hand holding a test tube with a QR code

Data platforms and insights

The Premonition Data Platform is a cloud-first platform for storing and analyzing biological and environmental data. It is designed from the ground up to integrate streaming signals (e.g. real-time sensor data), genomic data, and other contextual datasets (e.g. climactic, geospatial, domain-specific ontologies) into a versioned and traceable knowledge graph for cloud-scale analysis. It supports industry-grade IoT connectors for streaming sensor data, and its flexible APIs make it easy to capture data from laboratory workflows.

photo of a green-lit containment unit for lab testing

The Proving Ground

The Premonition Proving Ground is a unique biocontainment facility and habitat simulator for testing and training Premonition systems on disease vectors and other species with high economic impacts. Imported specimens are raised under accurate simulations of their native habitats, and then their natural responses to AI systems are observed. The Premonition Proving Ground can produce a terabyte of novel data every 24 hours — enabling rapid and cost-effective engineering and AI development.

close up of a mosquito and larvae

Example use-cases

close up photo of a lab technician filling a test tube

Biological early warning systems

Together with the National Science Foundation’s Technology, Innovation and Partnerships Directorate — and academic and public health partners — this program is developing scalable and adaptable biological early warning systems for state and local public health organizations. Pilots are proceeding in the Greater Houston area and are being led by Harris County Public Health. This program is initially focusing on cost-effective monitoring of mosquito-borne pathogens, which cause various diseases that can (re-)emerge in the United States.

Learn more
photo of a Nepalese woman standing next to a biological weather station

Monitoring neglected tropical diseases

Neglected tropical diseases (NTDs) impact more than a billion people worldwide, and many are transmitted by insects and other animals. Reducing their risk means understanding their prevalence and spread across remote and unmonitored locations. This program, in partnership with GSK and local partners in Nepal, is deploying Premonition systems into impacted communities far from the infrastructure found in major cities. The goals are to improve monitoring of mosquito-borne NTDs so essential medicines can be delivered to communities and to understand the requirements of sustaining advanced bio-monitoring capabilities in key ecologies.

Learn more
visual showing host pathogen data

Pathogen discovery with genomics

60-75% of emerging pathogens are zoonotic, meaning they jump from animals to people. Genomic sequencing has become a proven tool to help identify potential biothreats before outbreaks occur, by accelerating the discovery of novel viruses and microbes in the environment and helping to estimate their animals hosts. Our academic partners are using Premonition metagenomics AI to decode genomic data and discover potentially new pathogens in the environment, animals, and in clinical settings.

Learn more