What Changed at Corti

Corti's ambient documentation and text generation are now built into PATFAK version 6.6, Redline Data's software for German healthcare and social care facilities, following a pilot. Therapy sessions recorded in the PLINK app are transcribed and summarized into progress notes, and the Reporting module drafts reports, care plans and applications from treatment records. Use is billed in credits, with a monthly free quota for each license seat.

Our read

Roughly 1,500 rehabilitation, counseling and social care facilities on PATFAK get AI documentation without bringing in a separate vendor. The free credit quota lets them try it before committing budget.

Corti's Guided Generation now passes user edited, user added and discarded facts to the model with explicit type metadata. Edited facts take precedence over the transcript where the two conflict, and discarded facts are excluded from the note even when the transcript mentions them. Previously, edits and discards were not reliably reflected in the generated note.

Our read

This closes a common gap in ambient documentation, where a clinician's correction is undone by the next draft. Developers building on Corti should retest note generation, since conflicts now resolve in the clinician's favor rather than the audio's.

Corti announced the General Availability of its Guided Generation endpoints for text generation. Concurrently, the Documents Classic endpoint was deprecated, with a final sunset date of January 18, 2027.

Our read

Developers and healthcare organizations using Corti's text generation APIs must migrate to the new Guided Documents endpoint before the January 2027 sunset date to avoid service disruption. The GA status of Guided Generation also signals production readiness for structured clinical note generation.

Corti launched the ICD-10-UK Coding Expert for its Agent Library, a specialized AI model component designed to assign accurate medical codes localized to the United Kingdom standard. The Expert plugs into Corti's Agentic Framework to automate or assist with clinical coding based on patient encounter data.

Our read

UK-based healthcare organizations and NHS trusts can now leverage Corti's AI agent framework to handle local medical coding natively. This reduces administrative overhead and minimizes billing and documentation errors by using the specific ICD-10-UK classification system rather than generalized or US-centric coding models.

Corti added three new pre-built Experts to its Agent Library: the DrugBank Expert for structured medication and interaction queries, the PubMed Expert for retrieving direct research evidence, and the Interviewing Expert for running adaptive clinical questionnaires. These modules operate within the Corti Agentic Framework via the Model Context Protocol (MCP) to provide identifier-grounded intelligence without medical hallucination.

Our read

Healthcare developers can now easily build AI agents that tap directly into authoritative drug databases and peer-reviewed literature. This modular approach accelerates development for complex workflows like medication reconciliation while ensuring agents rely on verifiable, structured data rather than speculative generative models.