Built on the only model benchmarked safe enough for patient-facing care, Polaris 6.0 Ultra brings a 7.4-trillion-parameter safety constellation, a

Helps AI find the right pages in long documents by giving it richer clues than raw text alone.

OhioHealth partnered with Hippocratic AI to pilot a generative AI healthcare agent for pre-charting Medicare Wellness Visits, testing whether AI

Teaches voice AI to hear a patient’s cough as a clinical signal instead of background noise.

Nearly one in three older adults who reported no falls in the past year still showed two or more functional

Failure to titrate medications to guideline-recommended target doses is widespread across sixteen chronic conditions and carries substantial clinical and economic

Nearly half of the older adults who described their own health as good or better showed two or more mobility

A unified framework for understanding on-policy self-distillation as dense teacher-guided supervision in RL-based LLM post-training.

Most speech AI models can't respond while you're still talking — they lose track of the audio as it arrives,

How we engineered a voice stack that answers in under a second — and feels human doing it.

New Skills and Features for Healthcare Introduced: Built on more than 180m real patient interactions, validated by U.S.-licensed clinicians and

Tests whether AI agents can actually do real, paid professional work — not just pass toy benchmarks.

Why holding a natural healthcare conversation is one of the hardest problems in real-time AI.

A simple inference add-on that lets AI models handle long conversations with far less memory.

Patients and eye doctors gave high marks to a voice AI that teaches patients about eye-injection treatment.

A way to install a new behavior in an AI model — and cleanly switch it off later.

An academic-industry partnership delivering 11 short AI courses to 478 nursing students and faculty produced measurable knowledge gains alongside satisfaction

Announcing Polaris 5.0 — built on 180M+ real patient interactions and benchmarked against every major frontier model.

Makes the newest class of large AI models run significantly faster on the same hardware.

Helps AI answer complex questions by pulling knowledge from many angles at once instead of one.

Each memory the AI care agent carried over from a previous call was associated with roughly 2.5 more minutes of

Teaches AI models to keep the right memories — and drop the rest — during long conversations.

What 115 million real patient conversations taught us about making healthcare AI safer and more human.

The first benchmark measuring whether AI can genuinely emotionally support someone — scored side by side with humans.

Teaches AI to think in shorthand instead of essays — same answers, dramatically faster.

Grades AI agents not just on getting the right answer, but on doing the work the right way.

Posted by Amy McCarthy DNP, RNC-MNN, NE-BC, CENP Chief Nursing Officer at Hippocratic AI

Posted by Amy McCarthy DNP, RNC-MNN, NE-BC, CENP Chief Nursing Officer at Hippocratic AI

Meenesh Bhimani, MD, MHA - Chief Medical Officer
Amy McCarthy DNP, RNC-MNN, NE-BC, CENP Chief Nursing Officer

Meenesh Bhimani - Chief Medical Officer

Meenesh Bhimani - Chief Medical Officer

Spotlight Series with Co-founder Kim Parikh Interviewing Michelle Voisard - Director of Care Management

Shows how attackers trick AI over the course of a conversation — and how to stop them.

Spanish-speaking patients called by a multilingual AI care agent opted in to colorectal cancer screening at 2.6 times the rate

A Spanish-speaking AI caller got 2.6x more patients to say yes to cancer screening — AI closing care gaps, not

AI can extend nursing capacity across decision support, monitoring, and documentation, but only where nurses help shape design and implementation.

How AI agents are rewriting the way software gets built — a view from our CPO.

A new safety standard: licensed clinicians stress-testing healthcare AI on real outputs at scale.

When 6,234 licensed clinicians reviewed more than 307,000 real patient calls, the rate of clinically correct guidance rose from about

A new safety standard: thousands of licensed clinicians stress-testing healthcare AI before it reaches patients.

The definitive guide to connecting AI models with structured knowledge — with Meta, Amazon, and Adobe researchers.

Our core safety architecture — the real-time feedback loop behind every call — is now a granted US patent.

Meenesh Bhimani, MD, MHA - Chief Medical Officer - Hippocratic AI
R. Hal Baker, MD, FACP - SVP &

Vishal Parikh - Chief Product Officer

The case study behind our JMIR paper: 2.6x higher cancer-screening opt-in among Spanish speakers.

Announcing Polaris 2.0: a 3-trillion-parameter constellation that made healthcare AI safer, smarter, and multilingual.

Posted by Subhabrata Mukherjee, PhD

Real stories of the AI catching problems on calls and getting patients to the right level of care.

Making an AI voice that doesn’t just say the right thing — it sounds like it cares.

Posted by Dr. Meenesh Bhimani, MD Hippocratic AI

Michelle Voisard - Director of Care Management

Shanil Puri - Director, Speech Technologies

By Kriti Aggarwal - Staff Applied Scientist

Akash Chaurasia - Product Research Scientist

How the agent uses motivational interviewing and bedside manner to build trust and change behavior.

How the agent uses motivational interviewing and bedside manner to build trust and change behavior.

Small personal touches — calling at the right time, on the right channel — measurably improve health.

How the agent adapts its personality to each patient — because empathy isn’t one-size-fits-all.

Cuts AI costs by sending easy questions to a small model and saving the big model for hard ones.

The original blueprint for safe healthcare voice AI: a team of specialist models collaborating and checking each other’s work.