Collaborating on a nationwide randomized study of AI in real-world virtual care

2026-02-03 · Source: The latest research from Google · Field: Health & Wellbeing — Medical Devices & Health Technology, Clinical Care & Medical Practice, Public Health & Epidemiology · Depth: Intermediate, short

Summary

Google Research, in partnership with Included Health, is launching a nationwide randomized study to evaluate conversational AI in real-world virtual care workflows, pending Institutional Review Board (IRB) approval. Scheduled for February 3, 2026, this study aims to gather rigorous prospective evidence on AI performance in clinical settings at scale, moving beyond simulations and retrospective data. This research builds on foundational work with systems like AMIE for diagnostic and management reasoning, the Personal Health Agent (PHA) for personalized health insights from wearables, and a "wayfinding" AI for navigating health information. The initiative follows earlier studies, including a single-center feasibility study with Beth Israel Deaconess Medical Center, which showed strong indications of safety. This large-scale randomized controlled trial will compare AI-managed patient interactions against standard clinical practice to understand capabilities and limitations for real patients and concerns.

Key takeaway

For research scientists developing AI for healthcare, this study sets a new standard for evidence generation, moving from simulated environments to large-scale, real-world randomized controlled trials. You should prioritize rigorous, phased evaluation of AI systems, focusing on safety and utility in actual clinical workflows. This approach is essential for building trust and ensuring responsible deployment of AI-powered care, making high-quality medical expertise more accessible.

Key insights

A nationwide randomized study will rigorously evaluate conversational AI in real-world virtual healthcare settings.

Principles

• Rigorous evidence generation is crucial for medical AI.
• Phased research ensures safety and usefulness.
• AI can enhance access to medical expertise.

Method

The study employs a randomized controlled trial design with consented participants nationwide, comparing AI-managed patient interactions to standard clinical practice in virtual care workflows.

In practice

• AMIE: AI for diagnostic and management reasoning.
• PHA: AI for personalized health insights from wearables.
• Wayfinding AI: AI for navigating health information.

Topics

• Conversational AI
• Virtual Care
• Healthcare AI
• Randomized Controlled Trials
• Medical Diagnostic Reasoning

Best for: Research Scientist, AI Researcher, AI Scientist, Data Scientist

Related on AIssential

• Exploring the feasibility of conversational diagnostic AI in a real-world clinical study — Supervised conversational AI for pre-visit history taking is feasible, safe, and well-received by patients and clinicians.
• From Radiology to Drug Discovery, Survey Reveals AI Is Delivering Clear Return on Investment in Healthcare — Healthcare and life sciences are rapidly moving from AI experimentation to execution, driven by strong ROI and open-source adoption.
• A randomized controlled trial of artificial intelligence-based analytics for clinical deterioration — Passive AI-based predictive analytics displays did not significantly improve patient outcomes in a clinical trial.
• Vivek Natarajan of Google DeepMind at RAAIS 2026 — Building useful AI in expert domains requires moving beyond benchmarks to real-world clinical interaction and new knowledge generation.
• DeepER-Med: Advancing Deep Evidence-Based Research in Medicine Through Agentic AI — DeepER-Med enhances medical AI trustworthiness via explicit evidence appraisal and agentic collaboration.
• Decentralized Trials That Scale Globally Without Compliance Risks - with Emma Vitalini of Amgen — AI and decentralized trials accelerate patient recruitment and enhance compliance in regulated healthcare.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The latest research from Google.