Samwise Healthcare IT Newsletter
Monday, July 20, 2026
Trump Administration Launches Gold Eagle AI Initiative to Detect and Fix Cyber Vulnerabilities
The White House launched Gold Eagle, an AI-powered initiative to detect and patch cyber vulnerabilities at speed and scale never seen before, per a June 2 executive order on AI security. National Cyber Director Sean Cairncross and DHS Secretary Markwayne Mullin announced the program, which enlisted Treasury and CISA in its initial rollout. Gold Eagle operates as an interagency effort designed to identify exploitable weaknesses across critical infrastructure—including healthcare networks—and coordinate rapid remediation. The initiative marks the Trump administration’s most concrete federal commitment yet to AI-driven cyber defense, with vulnerability prioritization already under way as of this week’s announcement.
Sources: Healthcare IT News Share ↗ ✉︎ Email 💬 Text
CHAI Launches PULSE to Accelerate GenAI Adoption Across 2,000 Public Health Agencies
The Coalition for Health AI launched PULSE, a national initiative to bring generative AI tools to 2,000 public health practitioners who shape disease surveillance, emergency response, and community health policy. The program will pilot five genAI use cases with enterprise licenses donated by Anthropic and OpenAI—each supporting up to 2,000 seats—with onboarding managed by Accenture. Applications close August 7 and pilots begin fall 2026, with implementation playbooks slated for 2027. The initiative responds to research showing roughly 40% of local health departments are not yet using AI, a gap CHAI advisor Dr. Ashish Jha called unsustainable given the public health demands ahead.
Sources: Healthcare IT News Share ↗ ✉︎ Email 💬 Text
AMA Convenes Initiative to Map SNOMED CT to CPT Codes, Enabling FHIR-Based Prior Authorization
The American Medical Association is convening healthcare organizations to develop a SNOMED CT-to-CPT code mapping designed to enable FHIR-based prior authorization, with a January 1, 2027 compliance deadline approaching. The initiative calls on payers, EHR developers, clearinghouses, prior-authorization vendors, and standards bodies to collaborate on a bridge between clinical terminology and billing codes, aligning with Da Vinci implementation guides and CMS interoperability priorities. AMA President Dr. Willie Underwood III is leading the charge, asking organizations to contact the AMA directly to participate. The mapping is viewed as a critical prerequisite for automating prior-authorization workflows that today consume enormous administrative resources across the health system.
Sources: Healthcare IT News Share ↗ ✉︎ Email 💬 Text
Luma Health's 'Headless AI' Concept Aims to Eliminate Healthcare Bottlenecks Without a User Interface
Luma Health cofounder and president Aditya Bansod is making the case for “headless AI”—autonomous agents that operate as a coordination layer across healthcare applications without their own user interface. Where traditional AI tools surface inside a specific app, headless agents would navigate EHRs, payer portals, and scheduling platforms on behalf of staff, handling prior authorization, referral management, and routine administrative tasks. Governance, Bansod argues, must follow each AI agent through role-based access controls and complete audit trails enforced at both the application and platform level—a challenge he believes the industry has not yet fully addressed.
Sources: Healthcare IT News Share ↗ ✉︎ Email 💬 Text
AI and Human Hybrid Cuts No-Shows 24%, Front-Office Costs 50% at Graybill Medical Group
Graybill Medical Group, a 94-year-old physician-led community practice, cut front-office costs by 50% and reduced patient no-shows by 24% after layering AI into its communication workflows—but only after first mapping the operational problems driving poor access. Rather than buying software hoping for improvement, Graybill partnered with Third Way Health, a vendor combining AI agents with human expertise, and had the vendor learn existing processes before introducing any technology. Staff retained responsibility for empathy-dependent interactions while AI handled scheduling, FAQ routing, and appointment confirmation. COO Jamie Reddick’s advice to other practices: fix the operational problem first, then add AI to support the solution.
Sources: Healthcare IT News Share ↗ ✉︎ Email 💬 Text
OpenEvidence Partners With Boston Children's Hospital to Study AI's Impact on Clinical Care Patterns
OpenEvidence, the AI medical information platform now valued at $12 billion following a $250 million Series D in January 2026, announced a multi-year research collaboration with Boston Children’s Hospital to study how physicians’ use of AI at the point of care influences clinical practice patterns. OpenEvidence will be integrated directly into Boston Children’s EHR, giving providers access to its evidence-based AI copilot—trained on peer-reviewed medical literature—without leaving their workflow. The partnership will analyze anonymized data on how clinicians query the system during patient encounters; no patient data will be accessed. Chief Innovation Officer John Brownstein said the collaboration aims to surface health trends hidden in clinical practice.
Sources: MobiHealthNews Share ↗ ✉︎ Email 💬 Text
Hebrew University's New Genetic Programming Platform Could Advance Precision Medicine
Researchers at Hebrew University of Jerusalem published a new synthetic gene circuit platform in Nature this week that reduces the number of genetic components needed to perform complex computations inside human cells—a development that could enable future therapies to be programmed like software. The platform demonstrated biological equivalents of computing components including half adders, full adders, and multiplexers, while producing Interleukin-15, a cancer-fighting immune protein, in human embryonic kidney cells. By requiring fewer computational layers, the system lowers cellular resource demands that have historically made gene circuits impractical at scale. The researchers said the toolkit supports precision diagnostics, targeted therapies, and biosensing applications.
Sources: Healthcare IT News Share ↗ ✉︎ Email 💬 Text
Hemispheric Raises $52M to Build AI Brain Foundation Model for Mental Health Diagnosis
Hemispheric, a NeuroAI startup, emerged from stealth this week with $52 million in early-stage funding to commercialize a foundation model for the human brain. Its Descartes model—six billion parameters trained on more than 250,000 hours of brain activity data—analyzes EEG signals from a non-invasive dry headset worn during a 15-minute session to generate objective diagnostic profiles for clinicians. Cofounded by computational neuroscientist Hagai Lalazar and Gidi Littwin, co-inventor of Apple’s FaceID technology, Hemispheric is targeting PTSD, traumatic brain injury, depression, Alzheimer’s disease, schizophrenia, and anxiety—conditions that currently lack objective lab tests. The company is in regulatory discussions and planning U.S. and European market expansion.
Sources: MobiHealthNews Share ↗ ✉︎ Email 💬 Text
Curated by JD · samwise.agency
Leave a Reply
You must be logged in to post a comment.