AWS announces further initiatives in GenAI for public sector, precision healthcare
By Yogesh Hirdaramani
A US$50 million fund from Amazon Web Services (AWS) will provide public sector customers with credits to build new GenAI solutions.
In his opening keynote, AWS VIce President of Worldwide Public Sector, Dave Levy, shared on how AWS is supporting government agencies in adopting GenAI. Image: Amazon Web Services
At Amazon Web Services (AWS) Summit at Washington DC, generative AI (GenAI) dominated the cloud service provider’s public sector offerings.
AWS announced its Public Sector Generative Artificial Intelligence Impact Initiative, a US$50 million (S$68 million) fund that would be available for new and existing global public sector customers.
The initiative will allow public sector organisations to tap on AWS promotional credits to build new solutions using AWS’ GenAI tools, such as Amazon Bedrock, which provides access to foundation models, Amazon Q, a GenAI powered coding assistant, and Amazon Sagemaker, a platform that enables the creation, training, and deployment of AI tools.
They will also have access to training and technical expertise across these projects.
According to an April report by Forbes, AWS is a core growth engine for Amazon, with high demand in GenAI services driving growth.
GenAI a seismic shift poised to reshape society
In his keynote address, AWS Vice President of Worldwide Public Sector, Dave Levy, characterised GenAI as a seismic shift that is “poised to redefine the way we work, transform every customer experience we know, and shape some of society’s greatest challenges”.
The event, held from June 26 to 27, saw technology leaders from AWS, along with AWS customers and partners, convene to discuss GenAI, cloud services, and citizen services. Apart from GenAI, AWS announced increased investments in precision medicine.
In the US, federal agencies such as the Food and Drug Administration (FDA)’s Centre for Devices and Radiological Health use AWS’ GenAI services, such as natural language processing and intelligent document processing to review medical device reports, said Levy during his keynote.
Driving precision medicine for children’s health
AWS also announced an investment of US$10 million to support research into rare diseases amongst children.
Of this amount, the hyperscaler will be committing US$3 million to three organisations, with each receiving US$1 million, to accelerate data-driven research efforts into rare diseases. These organisations include the Children’s National Hospital in Washington DC, the Nationwide Children’s Hospital in Columbus, Ohio, and the Children’s Brain Tumor Network located within the Children’s Hospital of Philadelphia (CHOP).
Another US$7 million will be available for organisations working on pediatric research through the AWS IMAGINE Grant: Children’s Health Innovation Award. Eligible projects include research on rare diseases that children face, such as pediatric cancers, heart conditions, and genetic disorders.
Healthcare data discovery platform
During a presentation, CHOP’s Director of the Center for Data Driven Discovery in Biomedicine, Adam Resnick, announced that the team is launching a new platform titled D.A.V.I.D., or Data, Analytics, Visualisation, Insights and Discovery, in partnership with AWS.
D.A.V.I.D. will be a GenAI-assisted platform that clinicians would be able to use to ask questions and retrieve data on rare diseases. This builds on CHOP’s pre-existing partnership with AWS to enable hospitals to share data on rare childhood diseases securely with each other through Amazon HealthLake.
“It still surprises me that very few clinicians can answer the following question: Show me all the other patients at my institution that have had this disease with this mutation, what treatments did they get, what were the outcomes?” he said.
He explained that most brain tumours amongst children result from rare diseases with few off-the-shelf solutions. This is why data sharing amongst organisations was critical to finding precision health solutions for rare mutations.
AI services to support landmine clearance
AWS will provide a US$4 million package to HALO Trust, a non-profit organisation that focuses on clearing landmines in over 30 countries, for a pilot project that uses AI to detect debris of war, such as landmines and other explosives, in Ukraine.
The organisation currently uses high-resolution drone imagery to survey Ukraine’s minefields and detect areas that workers or machines need to clear. AI will speed up the process if the pilot is successful.
So far, HALO Trust has flown 542 drone flights over Ukraine’s minefields, and collected over 11 terabytes of data.
“[The pilot] has the potential to step-change how we identify minefields and unexploded bombs, so that we can better target clearance efforts and save more lives. That will allow us to work with the ambition and scale that is required to tackle what is a huge problem,” said HALO Trust Head of Development, Simon Conway, at a press conference at the summit.
AWS will also help HALO automate its use of satellite imagery to more quickly detect buildings damaged by explosives as well as signs of humans of activity.