Sponsored by: Hyro

94% of healthcare leaders call AI agents critical to operations. Only 59% have a standardized way to measure how they're actually performing.

Hyro's 2026 Healthcare AI Agent Benchmark Report, based on insights from nearly 400 health systems, introduces the first real-world benchmarks for AI agent performance in healthcare call centers. Download it free here.

Good morning, health AI enthusiasts —

Let’s dive in.

  • AI catches what humans miss

  • NYC hackathon recap

  • 3 in 10 US adults use AI for health info

  • 33 new tools/partnerships, 11 funding updates, new AI jobs & link-worthy content

Read time: 6 minutes

Our Picks

Highlights if you’ve only got 2 minutes…

1/

AI catches what humans miss

Two stories this week highlighted one of AI's biggest opportunities in healthcare: helping patients and clinicians catch things that might otherwise slip through the cracks.

The first came from health AI founder of Keragon, Conno Christou (a friend of the Healthcare AI Guy team) who documented how he used Claude throughout his treatment for a rare form of non-Hodgkin's lymphoma. After receiving conflicting recommendations from two leading oncologists, he gathered 12 medical opinions and used AI to analyze his scans, bloodwork, wearable data, and treatment journey. Near the end of treatment, Claude even flagged a common but often overlooked cause of false-positive PET scans, helping lead to the correct diagnosis and avoiding unnecessary radiation. Glad he’s better now!

The second came from NewYork-Presbyterian, where an AI model called EchoNext identified severe heart damage from a routine ECG after a patient had been sent home with asthma medication. Follow-up testing revealed a rare genetic heart condition, ultimately leading to a life-saving transplant. The FDA-cleared model will now be available through OpenEvidence, putting it in the hands of roughly half of US doctors.

Neither story is about AI replacing doctors. They're about pattern recognition at a scale humans simply can't match. Whether it's a patient asking better questions or a clinician getting another set of eyes, AI is becoming a second opinion that's available all the time.

2/

NYC hackathon recap

This past weekend we hosted another NYC Healthcare x AI Hackathon, bringing together an incredible group of eager builders. Over 15 teams of engineers, founders, clinicians and operators spent 1.5 days building AI solutions for healthcare. Huge thank you to everyone who applied, joined us, and to Baseten for hosting us in their NYC office after raising $1.5B last week at a $13B valuation. They power some of the most meaningful companies in health AI, including OpenEvidence, Abridge, EliseAI, and Tennr, so it was awesome to be the first partner to host a hackathon in their space.

The future of healthcare will be AI-native, deeply personalized, and increasingly patient-centric. It was inspiring to spend the weekend with people building toward that future. Here were the winners:

  • 🥇 Parsel: Parsel is a Parkinson’s progression tracker that uses voice as a biomarker to detect subtle changes in pitch, cadence, and tremor against each patient’s baseline. Combined with conversational AI check-ins, it turns a patient’s phone into a continuous monitoring tool between visits. (link)

  • 🥈 ContinuaCare: ContinuaCare built an agentic voice AI that automates Medicare’s Transitional Care Management workflow, from post-discharge outreach and risk screening to scheduling follow-up visits and helping practices capture TCM reimbursement that often goes unclaimed.

  • 🥉 SafeSight: SafeSight is an AI vision system for care settings that detects potential safety risks in real time, alerts care teams, and automatically generates incident reports to improve patient safety and clinical response.

Thanks to all our sponsors, judges, and co-hosts! See you all at the next one. (linkedin)

3/

3 in 10 US adults use AI for health info

People are turning to AI for health advice much faster than expected. A new KFF survey found that 29% of U.S. adults now use tools like ChatGPT, Gemini, and Claude for health information at least once a month, up from 17% just two years ago. That's now nearly the same as the 31% who turn to social media.

The reasons are pretty practical. People want answers immediately, they want privacy, and many are trying to figure out whether they actually need to see a doctor. Nearly 1 in 5 also said healthcare costs or difficulty accessing care pushed them toward AI instead. Makes sense.

One stat that stood out: 42% of people who use AI for health questions never follow up with a clinician. That should probably make the healthcare industry pause. Whether we like it or not, AI is becoming the front door to healthcare for millions of people, and that shift is happening a lot faster than I think most expected. (link)

TOGETHER WITH FRESHWORKS

Keep clinicians focused on patient care with Freshservice—AI-powered ITSM that reduces downtime and automates IT workflows

Tools & Partnerships 🔧

Latest on business, consumer, and clinical healthcare AI tools and partnerships…

TOOLS

  • UpDoc earns FDA clearance for AI diabetes tool: The clinical AI startup received FDA clearance for an AI system that helps titrate insulin doses and announced $18M in seed funding alongside a Cleveland Clinic pilot. (link)

  • US restores access to Anthropic’s Mythos 5: The government lifted restrictions on Anthropic’s flagship AI model, allowing deployment to more than 100 trusted U.S. institutions after the company implemented additional safeguards. (link)

  • Boston Children’s saves 30,000 hours with AI: The hospital has deployed more than 50 automation initiatives across the organization, generating over 30,000 hours in time savings during the first half of 2026. (link)

  • Kaiser cuts MRI wait times with AI: Kaiser Permanente reduced MRI wait times by more than 60% using an FDA-cleared AI tool that shortens scan times while maintaining image quality. (link)

  • Ambience expands AI tools for nurses: The company introduced new nursing-focused AI capabilities that automate shift preparation, documentation, and structured flowsheet completion. (link)

  • NVIDIA launches BioNeMo Agent Toolkit: The new toolkit equips AI agents with tools for protein structure prediction, molecular docking, and generative chemistry to accelerate drug discovery workflows. (link)

  • Neko Health debuts next-generation AI health scan: Ahead of its U.S. launch, the company unveiled Gen 2, adding body composition analysis, wearable integration, and personalized health insights to its AI-powered preventive screening platform. (link)

  • Redox expands healthcare AI infrastructure: Redox launched new AI capabilities, including an MCP server and AI Assistant Suite, to help healthcare organizations build and scale AI-powered workflows. (link)

  • Infinitus launches AI triage safety system: The company introduced Clinical Escalations, a risk detection system that continuously monitors AI conversations to identify urgent clinical situations and help prevent patient under-triage. (link)

  • OpenAI unveils GPT-5.6 model family: OpenAI introduced GPT-5.6 Sol, Terra, and Luna, with Sol becoming its most capable model yet while remaining limited to vetted partners as the company works toward broader availability. (link)

  • UnitedHealthcare expands AI assistant Avery: UHC is rapidly scaling its AI assistant to help members estimate costs, find in-network providers, and schedule appointments, with eligibility expected to grow from 7M to 21M users this year. (link)

  • Study finds emotional intelligence gap in voice AI: Researchers from Stanford and Together AI found leading real-time voice models often recognize emotional cues but fail to act on them appropriately in high-stakes healthcare scenarios. (link)

  • SimplePractice launches AI assistant for therapists: The company introduced Care Aide, an AI assistant that supports mental health clinicians with session preparation, treatment planning, intake review, and documentation. (link)

  • JAMA warns of hidden risks in clinical LLMs: A new viewpoint argues that biases and errors embedded in EHR data could lead LLMs to make unsafe clinical recommendations, calling for stronger privacy protections and greater transparency. (link)

  • Nabla Bio unveils next-generation AI drug design model: JAM-2 expands zero-shot drug design to multispecific antibodies and intracellular targets, generating drug-grade candidates with atomic-level precision in as little as six weeks. (link)

  • Stanford researchers introduce Proto for AI biology: The new open-source framework lets researchers combine dozens of AI biology models into unified pipelines, simplifying complex workflows across DNA, RNA, proteins, and drug design. (link)

  • TigerConnect adds AI-powered scheduling: TigerConnect expanded its clinical communications platform with AI scheduling tools that automate staff scheduling, detect coverage gaps, and optimize workforce assignments. (link)

  • Invoca launches AI patient engagement agent: The company introduced Nico, an AI agent that personalizes patient communications across calls, messaging, and digital channels to improve engagement and convert more inquiries into appointments and revenue. (link)

PARTNERSHIPS

  • OpenEvidence + Pathway Labs: OpenEvidence partnered with Pathway Labs to integrate an FDA-cleared AI algorithm that detects structural heart disease from ECG images into its clinical decision support platform. (link)

  • Innovaccer + AWS: Innovaccer and AWS signed a multi-year partnership to help health systems and payers deploy enterprise-scale agentic AI using AWS infrastructure. (link)

  • Mount Sinai + Signal 1: Mount Sinai partnered with Signal 1 to deploy an AI governance platform for centralized oversight and monitoring of predictive, generative, and agentic AI models. (link)

  • Stripe + Anthropic + OpenAI: Stripe, Anthropic, and the OpenAI Foundation joined the $500M Intercept initiative to fund technologies aimed at preventing respiratory infections. (link)

  • Houston Methodist + HealthLeap: Houston Methodist deployed HealthLeap’s AI platform to identify nutrition risk across all inpatients and enable earlier clinical intervention. (link)

  • Hartford HealthCare + TigerConnect: Hartford HealthCare deployed TigerConnect’s AI-powered scheduling platform to automate clinician scheduling, identify coverage gaps, and reduce administrative workload. (link)

  • Oracle Health + Theator: Oracle Health partnered with Theator to integrate AI-powered surgical video analysis into its EHR, automatically generating operative reports and surgical insights. (link)

  • Baystate Health + Oracle Health: Baystate Health expanded its partnership with Oracle Health to deploy a new EHR, Clinical AI Agent, and enterprise financial platform across its system. (link)

  • Baptist Health + Epic: Baptist Health deployed Epic’s AI Build Assistant to accelerate EHR development and streamline analyst workflows. (link)

  • Mayo Clinic + ARPA-H: Mayo Clinic joined ARPA-H’s PARADIGM initiative to develop AI-guided clinical tools that help rural healthcare workers perform medical procedures. (link)

  • Duke Health + Texas Health Resources + Cadence: Duke Health and Texas Health Resources partnered with Cadence to deploy AI-enabled chronic care management across their health systems. (link)

  • Grow Therapy + Stanford: Grow Therapy and Stanford launched a research partnership to develop evidence-based benchmarks for evaluating AI safety in mental healthcare. (link)

  • K Health + Hartford HealthCare: K Health and Hartford HealthCare expanded their AI-powered virtual care partnership to include endocrinology services. (link)

  • Stedi + ChatGPT: Stedi launched a native ChatGPT integration that allows providers to verify patient insurance eligibility in real time. (link)

  • Doctronic + Simple HealthKit: Doctronic partnered with Simple HealthKit to combine AI-powered triage with at-home diagnostics for automated virtual care. (link)

Deal Desk 💰

Spotlight on latest capital raises, M&A, and investments…

FUNDING

  • Assort Health, an AI agent platform for patient journeys, raised $120M in Series C funding at a $1.2B post-money valuation. Menlo Ventures led, joined by Lightspeed, Felicis, First Round Capital, Chemistry, Joe Montana, Tau Ventures, and Quiet Capital. (link)

  • Trase, a McLean, Va.-based developer of an AI operating system for highly regulated industries, raised $107M in seed funding. Bob Nelson and ARCH Venture Partners led the round. (link)

  • Cadence, a chronic disease management platform for older adults, raised $100M in Series C funding. Spark Capital led, joined by Thrive Capital, General Catalyst, Coatue, B Capital, Corewell Health Ventures, Memorial Hermann, and Duke Health. (link)

  • xCures, a healthcare data company, has closed a $46M Series B funding round, led by Innovius Capital. (link)

  • Coval, an SF-based provider of infrastructure for testing, monitoring, and evaluating autonomous voice agents, raised $28M in Series A funding. Norwest led the round. (link)

  • Upside, a healthcare-focused housing platform, raised a $20M series A funding.  Aquiline led the round. (link)

  • Xella Health, an SF-based women's health startup, raised $4.7M from Precursor Ventures, Capital F, and Ulu Ventures. (link)

  • ChemT, a Singapore-based startup focused on AI biomanufacturing, raised $4M in seed funding led by Wavemaker Ventures and Seeds. (link)

M&A

  • Gyde + Osborne Insurance Services: Gyde acquired Osborne Insurance Services to expand its AI-native brokerage platform across the ACA and Medicare markets. (link)

  • Withings + Biosency: Withings acquired Biosency to expand its respiratory monitoring and connected care offerings. (link)

  • Iambic AI + Bayer: Iambic AI partnered with Bayer to accelerate AI-powered drug discovery and development. (link)

as of 6/29/26

Other Relevant News 🔍

News, podcasts, blogs, tweets, resources, etc…

  • HHS advances autonomous clinical AI agenda (link)

  • Patients prefer healthcare providers' AI agents to public chatbots (link)

  • Rhode Island mandates AI scribe disclosure (link)

Join our free live discussion to learn how physician groups are using agentic AI to improve patient access, grow revenue, & drive ROI 👇

AI Job Opportunities 💼

Explore our AI Job Board or contact us to feature roles in our newsletter…

  • Account Director, Healthcare at OpenAI, an AI lab company

    $189 - 240K | SF, NYC, Seattle (link)

  • Research Scientist at OpenEvidence, a clinical evidence AI platform

    N/A | SF, Miami (link)

  • Strategic Finance, GTM at Basten, an AI inference company

    $190 - $210K | SF, NYC (link)

Visuals of the Week 📸

Funny memes, cool pics, and interesting data from around the web…

That’s it for this week friends! Back to reading — I’ll see you next week.

Stay classy,

— Healthcare AI Guy (X/Twitter | LinkedIn)

PS. I write this newsletter for you. So if you have any suggestions or questions, feel free to reply to this email and let me know

How was this week's newsletter? Tap your choice below👇

Login or Subscribe to participate


You may also like these