What challenges did Phagos face before AI implementation?

Antibiotic resistance causes millions of deaths annually and traditional antibiotic development is slow and costly, taking over 10 years. The manual process to identify effective bacteriophages is exponential and untenable due to the astronomical number of phage-bacteria interactions to test.

What was the impact of the AI solution?

Phagos shortened bacterial treatment development from 10 years to 2 months, showcasing a 100X improvement over traditional methods. The platform demonstrated success in animal farming, treating over half a million animals in France, with plans to expand to human phage therapy by 2030. The company has grown significantly, building unique microbial AI models and a proprietary genomic data platform on AWS.

Source

ExploringEvidence: Medium65/100

Phagos Uses Amazon SageMaker Generative AI to Revolutionize Bacterial Infection Treatment

Use case typeTherapeutics researchUpdated Jun 13, 2026

Phagos, a Paris-based biotech startup, is pioneering AI-powered phage therapy to combat antibiotic-resistant bacterial infections using generative AI models on Amazon SageMaker to match bacteriophages to specific bacteria. The challenge addressed is the inefficiency and slowness of traditional antibiotic development amid rising antibiotic resistance causing millions of deaths annually. Phagos developed predictive AI models trained on genomic data using Amazon SageMaker, and utilizes Amazon EC2 for computation, Amazon S3 and RDS for scalable and secure data storage, enabling rapid phage-bacteria matching. The AI platform reduced treatment development time from over 10 years to about two months, achieved 99.5% time savings in phage screening, and allowed deployment in animal farming with plans for human use by 2030.

Organization: Phagos
Industry: Healthcare
Location: France
Published: April 2026

Reported outcomes

100x

quantified impactOther quantified impact

Strategic outcomes

New product / capabilityAutomated phage screening platformSpeed & agilityAccelerated treatment developmentMarket & geographic expansionExpanded into animal farmingInnovation & cultureBuilt proprietary microbial AI models

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Quantified impact: 100 x increase

AWS Customer StoriesApr 29, 2026Customer storyInferred claimMedium evidence strength

Phagos shortened bacterial treatment development from 10 years to 2 months, showcasing a 100X improvement over traditional methods.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Phagos
Provider: AWS
Maturity: Exploring
Linked source: AWS Customer Stories

Amazon EC2 provides computational power for biological simulations, while Amazon S3 and RDS offer flexible, scalable, and secure storage for genomic and experimental data

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1Generative AI
2Healthcare AI
3Drug Discovery

Antibiotic resistance causes millions of deaths annually and traditional antibiotic development is slow and costly, taking over 10 years.
The manual process to identify effective bacteriophages is exponential and untenable due to the astronomical number of phage-bacteria interactions to test.

Phagos uses Amazon SageMaker to train generative AI models to predict phage-bacteria interactions and quickly identify optimal phages for treatment.
Amazon EC2 provides computational power for biological simulations, while Amazon S3 and RDS offer flexible, scalable, and secure storage for genomic and experimental data.
The AI-powered platform automates phage screening, reducing test counts by 50% and cutting screening time per bacteria from 29 hours to 10 minutes.

Technologies

Amazon SageMaker Amazon EC2 Amazon S3 Amazon RDS

Phagos shortened bacterial treatment development from 10 years to 2 months, showcasing a 100X improvement over traditional methods.
The platform demonstrated success in animal farming, treating over half a million animals in France, with plans to expand to human phage therapy by 2030.
The company has grown significantly, building unique microbial AI models and a proprietary genomic data platform on AWS.

Architecture

Phagos built its AI platform on Amazon SageMaker for generative AI modeling, uses Amazon EC2 for computational simulations, and stores data securely and scalably on Amazon S3 and RDS, integrating with laboratory management and external data systems.

Sources & evidence1

Evidence: Medium65/100Evidence strength