AI News

OpenAI 将计算机操作智能体接入医疗工作流

Unite.AI更新于 11月16日 23:10

OpenAI Brings Computer-Use Agents to Healthcare Workflows

AI 导读

OpenAI 于 11 月 16 日发布医疗计算机操作智能体,通过 ChatGPT for Healthcare、ChatGPT for Clinicians 和 OpenAI API 提供,并称在 325 个经审核的 GPT-6 Astra 电子健康记录收件箱任务中,99% 被评为临床安全。这些智能体操作现有软件界面完成授权申请、转诊和出院规划等多步工作,在演示中可起草 MRI 医嘱、就诊记录和患者回复供医生审核。OpenAI 还报告 GPT-6 Astra 在斯坦福 HealthAdminBench 上获得 94.3% 的评分点,GPT-6 Luna 为 91.8%,后者每任务成本 8.5 美分,比 Astra 低 98%。

推荐理由

精选AI 评分 82

原文给出医疗智能体在真实行政流程中的落地案例与安全评估口径,可据此判断计算机操作智能体的部署边界。

主题

标签

正文

OpenAI announced healthcare computer-use agents on November 16, 2026, making them available through ChatGPT for Healthcare, ChatGPT for Clinicians and the OpenAI API, and reported that 99% of 325 reviewed GPT-6 Astra tasks in electronic health record inboxes were rated clinically safe.

The agents operate existing software interfaces to complete multistep work, including authorization requests, referrals and discharge planning. In a synthetic-patient demonstration, a message about lower back pain prompts a review of visit notes, physical-therapy records and prior imaging. The agent drafts an MRI order, encounter note and patient reply for clinician review, then prepares an insurer-portal authorization request for an authorization specialist to review.

Safety and Cost Evaluations

The safety assessment examined whether inaccuracies, omissions or actions in the model’s work could plausibly lead to patient harm. Reviewers assessed Astra’s interface actions alongside the notes, orders and patient messages it produced. OpenAI also reports that physicians selected the two highest helpfulness categories, very helpful or extremely helpful, in 93% of 646 assessments on a five-point scale.

On HealthAdminBench, an independent benchmark from Stanford researchers, OpenAI reports that GPT-6 Astra earned 94.3% of available rubric points and GPT-6 Luna earned 91.8%. Luna’s cost of 8.5 cents per task was 98% lower than Astra’s, and both models ran at maximum reasoning effort.

The benchmark contains 135 tasks and 1,698 subtasks spanning prior authorization, durable medical equipment orders and appeals. Agents observe live web portals through screenshots, accessibility trees or both, then navigate forms, review clinical documentation, check coverage criteria and submit decisions through standard browser interactions.

OpenAI reports that a CVS Health production agent cleared a months-long document-sending backlog over a weekend, while another handles system steps for tens of thousands of specialist consults each day. The agents navigate existing applications to complete steps that application programming interfaces do not fully support. CVS Health uses validation checks, human oversight, screenshots and action records to verify actions and application state. Data remains within its enterprise-controlled infrastructure, and production data is not used to train foundation models.

CVS Health has also published its CUA Agent Kit under an Apache-2.0 license. The kit uses an OpenAI computer-use agent to discover browser or Windows workflows, then turns successful runs into inspectable Playwright or pywinauto code that runs without model calls. This removes repeated model reasoning from stable workflows, while Codex-assisted diagnosis and updates remain subject to engineer review, testing and verification.

Cigna Healthcare reports that its case managers previously navigated more than 10 separate clinical and administrative systems to gather benefits and coverage, provider, appointment and prior-interaction context. Its deployment assembles that information for member outreach and moved from concept to deployment in under eight weeks. The work is part of The Cigna Group’s broader collaboration with OpenAI and is intended to give case managers more time to support members around care plans established by treating providers.

Human Review and Access Controls

Abridge is developing a cross-system assistant that can coordinate discharge work across medication access and home health, with concurrent agents and escalation for clinical judgment. Its own announcement says the experience is live in early environments with a few select partners, and that oversight must show what information was used, whether each step completed, when the assistant escalated and whether a payer received a submission.

Ambience’s demonstration applies similar oversight to prior authorization. The agent prepares a request from chart evidence, stops to ask a clinician when the record is ambiguous and waits for approval before submitting. Health systems can restrict permitted actions, pause the agent and inspect a step-by-step audit log.

CVS Health’s repository warns that run artifacts are not sanitized exports: traces can contain credentials, cookies, session tokens and network data, while screenshots capture visible information. Credential replacement and redaction markers are best-effort protections rather than comprehensive redaction.

ChatGPT for Healthcare provides role-based access controls and audit logs, and organizations can add approval steps and limit agents to the applications, records and actions needed for each task. To support HIPAA-compliant workflows involving protected health information, organizations must use eligible services covered by an OpenAI Business Associate Agreement and the required security and retention settings. OpenAI states that organization data is not used to train its foundation models.

The contrast between CVS Health’s production administrative agents and Abridge’s early care-coordination environments defines the rollout’s central tension. Computer use is already handling high-volume interface work, while broader clinical coordination makes audit trails, access limits and escalation points essential to safe deployment.

来源

Unite.AI