Xuron

AI generates the virtual patient conversations, learner feedback and activity drafts that define Xuron's offering. Removing the models would remove the practice encounter and its automated review, rather than a supplementary feature.

Qualified personnel must approve content before deployment, and Assurance supports testing, human selection of changes and retesting. Live simulations and feedback can reach learners without prior review. Automatic stop thresholds and escalation conditions are unspecified.

Separate virtual human and feedback components process conversation inputs and evaluate them against course criteria. The underlying model families, external AI providers, training adaptations and model version history are not identified.

AI processing may involve outside providers under contractual data protection restrictions, and customer content cannot train third party models. Provider names and a complete model subprocessor inventory are unspecified, including retention terms after transfer.

A January 2026 Xuron authored Almanac evaluation covered 16 clinicians and 18 obesity counseling sessions. Knowledge and comfort improved after the combined course and simulation. The uncontrolled design cannot isolate simulation benefit, and automated assessments were not independently validated.

Customer content is excluded from training third party models, but transcripts are not automatically deidentified. A fixed retention period and first party training policy are unspecified. Assurance supports testing and review, without a complete safety incident response commitment.

The service privacy policy covers accounts, voice inputs and transcripts, including possible regulated information. Users carry responsibility for compliant submissions. Business associate status, BAA coverage and an explicit exclusion of all patient information remain unspecified.

Privacy controls include encrypted payment transmission, restricted personnel access and confidentiality requirements. These describe operational safeguards. An independent security attestation or a stated external penetration testing cadence is not established.

The terms limit Xuron to education and training, excluding use as medical advice or a substitute for clinical judgment. An FDA supported educational activity does not establish medical device clearance for the platform.

Assurance maps course context, evaluation, controlled changes and recorded decisions to the NIST AI Risk Management Framework. It links findings to course versions and reviewers. This is a governance workflow, without comparative subgroup results or NIST certification.

Assurance links feedback to the underlying conversation and records selected corrections and retests. Customers retain responsibility for content approval. Fees do not depend on AI accuracy, and no vendor obligation to remedy incorrect educational output accompanies the review workflow.

The platform creates learning activities, conversation feedback and review records. Named collaborators indicate educational distribution, but no named EHR connector, learning system interface or implemented interoperability standard accompanies the described workflow.

An internet connected service delivers the learning environment and stores institutional accounts and activity data. International transfers are permitted. A committed hosting country, deployment alternatives and tenant isolation arrangements remain unspecified.

Business subscriptions support monthly, quarterly or annual billing, with additional feature and usage charges. Current plan prices and included volumes are unspecified. The sales process begins with an institution's proposed activity and its implementation requirements.

Use cases include continuing medical education, academic medicine, workforce training and life sciences. Small evaluations cover obesity counseling and telehealth, while collaborators include West Virginia University and HealthStream. Those studies support only part of the broader application range.