US public health agencies to test OpenAI and Anthropic AI models

Public health departments across the United States will test generative AI tools under a new programme involving the Coalition for Health AI, OpenAI, Anthropic, and Accenture. The Public Health Use Case and Learning Scaling Engine, known as PULSE, will support trials in 10 state, local, tribal, or territorial jurisdictions. The programme is intended to produce implementation guidance for public health agencies considering similar deployments. OpenAI and Anthropic have donated 10 enterprise licences with capacity for up to 2,000 public health practitioners. Accenture will oversee participant onboarding and help develop playbooks based on the trials. The programme will give public health practitioners access to enterprise AI products from OpenAI and Anthropic. CHAI has not identified the products, model versions, or configurations that will be used, or explained how the two providers will be assigned across the pilots. “Every major technological transformation succeeds or fails based on trust, governance and execution,” Dr. David Lakey, a former Texas health commissioner, said. “PULSE will support agencies in this endeavour, and is specifically designed for practical implementation.” Five public health use cases The Coalition for Health AI’s leadership council will select the participating jurisdictions and assign practitioners to communities focused on five use cases. These include biosurveillance and drug-wave prediction, social determinants of health and SDoH mapping, operations and efficiency and community-feedback analysis, public communications and a multilingual translation hub, and automated clinical-data retrieval and a FHIR query engine. NIST’s AI Risk Management Framework recommends evaluating AI systems according to their intended use, operating environment, affected parties, and potential consequences. CHAI has not published separate evaluation, privacy, security, or human-review requirements for the five PULSE use cases. FHIR is an HL7 standard used to exchange healthcare information electronically between compatible systems. PULSE includes an automated clinical-data retrieval and FHIR query engine, although the announcement does not define the role of generative AI in that workflow. It does not state whether the models will generate queries, retrieve records, summarise returned information, or combine those functions. The announcement also does not explain whether incorrect queries, incomplete retrievals, or unsupported summaries will be checked by staff before the information is used. Some proposed applications could involve demographic, geographic, clinical, or population-health information. CHAI has not said whether the pilots will use identifiable records, de-identified information, synthetic data, or aggregated datasets. The US Department of Health and Human Services requires organisations covered by HIPAA to protect certain electronic health information. Its cloud-computing guidance states that regulated organisations and service providers must meet HIPAA requirements when cloud systems create, receive, maintain, or transmit electronic protected health information. HIPAA will not apply to every participating organisation or workflow. Its application will depend on the agency, the data involved, and the function being performed. OpenAI states that inputs and outputs from its business services, including ChatGPT Enterprise and its API, are not used to train or improve its models by default. Anthropic also says it does not use inputs and outputs from its commercial products for model training by default. Those provider policies do not define how the PULSE deployments will be configured. The announcement does not set out retention periods, access controls, audit arrangements, data-storage requirements, or rules governing the submission of protected health information. “We believe AI should be useful, safe and accessible to the people tackling society’s most important challenges,” Felipe Millon, OpenAI’s head of government go-to-market, said. He added that the licences were intended to help public health organisations evaluate the tools through a structured process. Governance and evaluation remain undefined The pilots are scheduled to begin in autumn 2026. CHAI expects the resulting playbooks to be released in 2027 and used as reference material by other public health agencies. CHAI has not published the measures that will be used to assess the pilots or explained whether each use case will be evaluated under separate technical, operational, privacy, and safety criteria. The announcement also does not explain how model outputs will be reviewed. It does not state whether staff must approve generated public communications, verify translations, validate retrieved clinical information, or review biosurveillance and drug-wave outputs before they are used. NIST guidance recommends identifying which AI functions require human oversight and training users to understand system performance and limitations. Its generative AI guidance also covers testing, validation, monitoring, documentation, privacy, and management oversight. “Public health teams are being asked to do more with less, and AI can help — as long as it’s brought in with care and the right guardrails,” Elizabeth Kelly, Anthropic’s head of beneficial deployments, said. She said PULSE would allow practitioners to test the tools in their own environments with privacy, governance, and responsible-use measures incorporated from the start. Data from the National Association of County and City Health Officials cited by CHAI showed that nearly 40% of local health departments were not using AI. The coalition said some departments were interested in revising workflows and improving operational efficiency. PULSE will provide enterprise licences, onboarding, peer communities, and implementation playbooks. CHAI has not specified the minimum staffing, infrastructure, interoperability, or cybersecurity requirements for participating jurisdictions. Eligible participants include state and territorial health departments, county and municipal agencies, tribal authorities, Indian health organisations, and large city health departments. PULSE plans to convert findings from 10 jurisdictions into guidance for wider use. The announcement does not explain how the playbooks will account for differences in agency size, technical systems, legal responsibilities, staffing, or procurement arrangements. The announcement also does not explain whether outputs from biosurveillance, drug-wave prediction, or clinical-data retrieval will be used only for testing, presented to staff for review, or incorporated into operational workflows. PULSE forms part of the Coalition for Health AI’s broader work on governance standards for healthcare AI. In May, the organisation announced plans to develop guidance covering eight governance areas through workshops and working groups involving more than 150 healthcare AI representatives. The coalition has since started publishing playbooks on organisational AI policies, governance structures, and internal resources. Additional guidance is expected to cover other areas of healthcare AI management. Separately, CHAI has worked with the Joint Commission on governance playbooks aligned with its voluntary Responsible Use of AI in Healthcare certification. The PULSE announcement does not state that participating public health agencies will be assessed under that certification. Dr. Brian Anderson, chief executive of the Coalition for Health AI, said public health agencies entered the COVID-19 pandemic after years of limited investment in technology. He said the PULSE programme was intended to give agencies practical experience with AI before wider implementation. “We know AI is going to reshape how we deliver public health — the question is whether we do it thoughtfully or not,” Dr. Ashish Jha, a former White House COVID-19 response coordinator, said. He said the programme would test which applications work and document the findings for other agencies. See also: Bunkerhill raises $55M to scale agentic AI across health systems Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security & Cloud Expo . Click here for more information. AI News is powered by TechForge Media . Explore other upcoming enterprise technology events and webinars here . The post US public health agencies to test OpenAI and Anthropic AI models appeared first on AI News .