Recursion Partners with Citeline to Integrate Real-World Data Into Its AI Drug Development Platform 🔗

Recursion Pharmaceuticals has signed a data partnership with Citeline to integrate real-world clinical intelligence directly into its AI-driven drug development platform. The collaboration expands Recursion's data inputs beyond its proprietary phenomics datasets, adding structured evidence on trial enrollment patterns, competitor pipelines, and clinical outcomes. Alongside the deal, Recursion appointed Vicki Goodman as Chief Medical Officer, bringing late-stage oncology development experience to a platform that has historically been strongest on the discovery side. The combination of richer data and clinical leadership is a deliberate move toward closing the gap between computational biology and clinical execution.

Insight Health Raises $11 Million to Scale Clinical AI Agents 🔗

Insight Health has closed an $11 million Series A to expand Lumi, its patient-facing clinical AI agent that conducts pre-appointment interviews, generates structured medical histories, and integrates them directly into EHR systems. The platform has already facilitated over 3 million autonomous patient interactions. Healthcare organizations using Lumi — including The Oregon Clinic and Santiam Hospital — report collective savings of over $50 million annually in administrative costs. The round was led by Standard Capital and includes integrations with platforms like athenahealth Marketplace. Clinical AI is moving from back-office automation into the patient-facing layer of care delivery.

GUIDE-AI Launches EU-Funded Initiative to Embed AI Guideline Navigation Into Clinical Practice 🔗

A new public-private initiative called GUIDE-AI, led by Charité – Universitätsmedizin Berlin and backed by €9.5 million from the EU's Innovative Health Initiative, is developing AI-driven therapy assistants that help clinicians navigate complex medical guidelines in real time. The system integrates large language models into hospital information systems to analyze individual patient data against current guideline recommendations — alerting physicians when prescribed treatment falls short of protocol for conditions including heart failure, COPD, chronic kidney disease, and asthma. A randomized clinical trial will evaluate whether the navigators produce measurably more guideline-compliant prescriptions. Nineteen partners across Europe and Israel are involved.

AI Is Becoming the Front Door to Medical Information — IQVIA 🔗

IQVIA published a detailed analysis documenting a structural shift in how healthcare professionals access clinical information. More than half of HCPs now use generative AI tools for scientific queries, with younger clinicians showing higher adoption still. The key finding: AI often precedes — or replaces — traditional pharmaceutical rep interactions as the first point of contact for clinical evidence. Platforms like OpenEvidence are gaining significant traction in the U.S. European adoption remains more cautious, shaped by the EU AI Act, while institutional-level deployments are accelerating in China.

The Signal: Pharma's commercial engagement model was built around the rep-to-physician relationship. That model is being disintermediated by algorithms. The credibility of the evidence that AI surfaces for HCPs is now a commercial variable — which means algorithmic positioning, evidence pipelines, and compliant AI content are no longer marketing functions. They are strategic infrastructure. Veeva's acquisition of Ostro two weeks ago and IQVIA's analysis this week are describing the same structural shift from opposite ends of the market.

Notch Raises $30 Million to Build an AI Operating System for Regulated Industries 🔗
Notch has raised $30 million in a Series A led by Headline, with Lightspeed Venture Partners participating, to build what it describes as an AI operating system for regulated sectors including life sciences. The platform combines conversational AI with deterministic workflow controls, compliance governance, and fully traceable outputs. Annual recurring revenue has grown 12x. The pitch is precise: general-purpose AI was not built for environments where every action must be auditable, defensible, and reversible on demand.

The Signal: The governance layer is becoming a product category with serious capital behind it. Notch is competing in the same space as Domino Data Lab's recent agentic lifecycle release and the traceability infrastructure being embedded into ELN platforms. The pattern is consistent — as AI agents scale into regulated life sciences workflows, the compliance wrapper is not a feature. It is the product. Companies solving governance first are accumulating structural advantages that bolt-on compliance cannot replicate.

Corti Launches Symphony — Agentic AI for Medical Coding Claims Accuracy Lead Over Big Tech 🔗

Copenhagen-based Corti has launched Symphony for Medical Coding, an agentic AI model it claims achieves over 25% higher accuracy in clinical coding benchmarks than systems from OpenAI, Google, and Anthropic. The model uses multi-agent architecture to reason through clinical evidence rather than pattern-match against annotated datasets, and is already deployed across more than 200 healthcare teams in the U.S. In a study using Danish patient data, Symphony tripled the recorded instances of suicide attempts compared to conventional coding — surfacing clinical evidence that existing systems had systematically missed.

The Signal: Medical coding sits at the intersection of clinical documentation, billing, and the population health datasets that train the next generation of AI models. A domain-specific agent that outperforms general-purpose models here is not a narrow win. It is an argument for vertical AI specialization across every documentation-heavy workflow in regulated healthcare. The accuracy claims will need independent validation — but if they hold, this category moves fast.