Clinical Reference & Evidence

FDB, formerly First Databank, supplies the drug knowledge that sits underneath most medication decisions made in the United States. Its databases are embedded in the majority of hospitals, physician practices and pharmacies, which means that when a prescribing system warns about an interaction, a dose ceiling or a contraindication, the content behind that warning is frequently this company's rather than the software vendor's. It is owned by Hearst and part of Hearst Health, the same group that owns Zynx Health. The asset is human curated: clinical editors maintain the content and the company's stated credibility rests on rigorous quality control and clinical expertise rather than on any model. What makes the record current rather than historical is a deliberate repositioning. In October 2025 FDB opened pilot integrations of a Model Context Protocol server, and on 31 March 2026 it made FDB MedProof MCP generally available, described as the first such server purpose built for agent driven medication decision support, letting other companies' AI agents and language models query clinically validated drug knowledge in patient specific context rather than answering from training data. Two further products were previewed in March 2026 and are not yet generally available: FDB Script Agent, which turns a spoken clinical conversation into a structured prescription for physician review, and FDB VerifyAssist, an inpatient pharmacy order verification assistant. The company is positioning itself as the grounding layer beneath other people's medication AI.

Synapse Medicine builds what it calls a medication intelligence platform, and the distinctive thing about it is where the machine learning sits. Rather than learning to make a prescribing recommendation, the company uses natural language processing to read and classify medication information from three kinds of unstructured source, manufacturer product documentation, official recommendations from health authorities, and published research, and to assemble that into a structured knowledge base that then drives interaction checking, prescription support and medication reconciliation. A separate product applies the same approach to pharmacovigilance itself, automating the coding and prioritisation of adverse drug reaction reports for the centres that receive them. The company was founded in Bordeaux in 2017 by Clement Goehrs and Louis Letinier, both physicians, with Bruno Thiao-Layel, and has raised roughly 39 million dollars across three rounds through a Series B in March 2022 led by Korelya Capital, with backing from a healthcare professionals' insurer and public investment banks rather than from pharmaceutical money. Its institutional position in France is unusual: users include the national medicines agency and the digital health agency alongside Assistance Publique Hopitaux de Paris, the Hospices Civils de Lyon and the Bordeaux teaching hospital, plus more than a hundred hospitals and dozens of telemedicine companies. The company states that it is completely independent of the pharmaceutical industry.

Clinical decision platform built inside Providence, a 51 hospital health system, and in use there since mid 2022. Founder and chief executive Dr Eve Cunningham, group vice president and chief of virtual care and digital health at Providence. The job it does is narrow and well chosen: supporting the transition between primary and specialty care, telling a clinician whether a referral is needed and, if it is, what workup should happen before the patient is seen. Guidance covers more than 730 conditions and is delivered inside the electronic health record, with the guidance mapped against that patient's own data so the clinician sees knowledge and patient context on one screen. Patient data surfaced from the record is stated not to be stored in the product, which is the second instance in this index of processing patient data without retaining it. The record is filed under decision support rather than clinical reference because it takes patient context and returns a patient specific next action, which is the test this index applies. It is cross listed to reference because the underlying asset is a curated guidance library. One scoping caution. The published account of how the platform was built, in a peer reviewed journal, states that all content and algorithms were built by clinicians in a proprietary environment requiring no code. Separate company and trade material describes generative artificial intelligence being integrated into the platform and calls it an artificial intelligence enhanced clinical intelligence engine. Nothing published reconciles those two descriptions or states what the generative layer does, and the capability notes grade accordingly. Corporate position is in transition and should be confirmed before contracting. Providence stated in late 2024 that it was preparing to spin the platform out, the peer reviewed paper carries an author disclosure about participation in commercialisation during 2024 and 2025, and the product website now carries a 2026 copyright notice in the name of a different entity.

Clinical assistant from Doximity, free to verified United States clinicians and reached inside the professional network the company has operated since 2010. Naming is currently in flux across the company's own material, appearing as Doximity GPT, DoxGPT and, in the support documentation, Doximity Ask. A separate record covers Doximity Scribe, the ambient documentation product; this record covers the assistant, following the product scoping rule that splits one company across records where the products serve different jobs. The evidence side has grown into the substance of the product. It answers clinical questions with referenced responses, gives direct access to peer reviewed literature, and lets a clinician select preferred journals and sources which are then prioritised in later answers, which is a degree of user control over the corpus that nothing else in this category offers. A peer reviewed drug reference covering more than 3,200 monographs with dosing, adverse effects and interactions is integrated alongside it. The company draws an explicit line inside its own product between where it retrieves and where it generates. Drug answers are stated not to be generated on the fly but returned from a structured peer reviewed dataset, and the material contrasts that with what it calls guesswork. Output is shaped for clinical reading, with conclusions placed first and comparisons presented as tables. The rest of the surface is documentation and correspondence: chart notes, discharge summaries, prior authorisation and appeal letters, patient education materials, multilingual translation, and coding assistance, with attachments for uploading labs or notes and saved projects for recurring templates. Secure messaging and fax sit alongside for sending material. Published commitments state encryption in transit and at rest, sessions private to the individual clinician, and prompt data never shared and never used to train the models.

All in one clinical artificial intelligence platform for clinics and hospitals, based at a San Francisco office address and operating since at least 2021, when trade coverage described the company. The product line is unusually wide for a company of this size: a clinical chat assistant, an artificial intelligence scribe, an artificial intelligence receptionist handling patient calls and scheduling, a real time interpreter across several languages, a decision support module offering diagnostic insights and treatment suggestions, a medical search engine, and a study companion for examination preparation. Free consumer style resources sit alongside them including a symptom checker, drug and calculator search, a question bank and contests. The company earlier operated a digital musculoskeletal and physical therapy application using computer vision motion tracking, and the current platform represents a change of direction rather than an extension of it. This record is scoped to the clinical evidence and search side, which is why it sits in this category, and is cross listed only to clinical decision support. The scribe and receptionist lines are described here rather than cross listed, because no evidence was located for any of them and propagating unverified product claims across further category pages would not be sound. The published claims require care and the capability notes set out why in each case. A homepage chart reports 99.4 percent on a named medical question answering benchmark against three named competitor models, with no methodology, date or source. A logo panel headed with a statement about being backed by world class technology leaders displays four major technology company marks, and a separate panel displays the marks of eight prominent academic medical centres above a count of professionals using the product. The nature of every one of those relationships is unstated, and none was verified.

Drug and toxicology reference from Merative, the company formed in June 2022 when Francisco Partners acquired the Watson Health assets from IBM. Headquartered in Ann Arbor, Michigan; Merative chief executive Gerry McCarthy, with Sonika Mathur as general manager for Micromedex. Content is curated by clinical experts from primary literature with in line referencing and daily updates, and is used in more than 80 countries by providers, poison control centres, government departments, universities, pharmaceutical organisations and health plans. The artificial intelligence layer is search. Released 23 September 2025, it lets a clinician ask natural language questions about drug information including interactions and intravenous administration, and returns answers drawn from Micromedex content with a citation on every result that opens the underlying source in one click. Interface affordances include suggested searches and follow up prompts. The company describes the capability as clinically validated. Scoping note, because the boundary matters for this index. DynaMedex is a separate joint product combining this drug content with EBSCO's DynaMed disease content, and it is graded on the DynaMed record rather than here; a December 2025 agreement additionally routes Micromedex dosing and medication safety content into that product's generative layer. This record covers Micromedex itself. Two facts sit outside what a reference product normally carries and both are on the record. Micromedex is recognised as a drug compendium under United States federal law, which makes its content a determinant of coverage for certain off label uses rather than merely a guide to them. And in a separate arrangement the company has licensed a subset of its drug content into a consumer artificial intelligence answer engine as a premium source. Named number one for point of care drug reference in the 2026 Best in KLAS report, for the second time.

Generative chat interface over the American Society of Clinical Oncology's own library of clinical practice guidelines, announced May 2025 and built with Google Cloud. Chief executive Clifford Hudis. Based in Alexandria, Virginia. Record scoped to the Guidelines Assistant rather than to the society, whose meetings, journals, research programmes and certification activities are outside what this index grades. The corpus is drawn solely from ASCO's published guidelines, which rest on systematic reviews and the work of expert oncologist panels, and is updated as guidelines are published or revised. Responses carry citations and sources, and a clinician can ask follow up questions. Access requires an ASCO account and is available to members through the society's site and its member application on both mobile platforms. Two things make this record structurally different from the rest of the category. The first is that the model layer is named openly: the tool is built on Google Cloud's Vertex platform using Gemini models, and no other vendor graded in this category names its model family or its provider. The second is the position the society occupies. ASCO authors the guidelines that define standard of care in medical oncology, operates a certification programme under which other vendors' oncology pathway products are assessed, and now also ships a product built on its own guidelines. The index already records a separate oncology pathways product carrying an ASCO designation. The published disclaimer is unusually precise and is carried on the record because it bounds the product properly: the assistant is described as a voluntary educational resource that does not account for variation among patients and does not substitute for the independent judgment of a licensed healthcare professional. User feedback is stated to refine both the tool and the development of the guidelines over time.

Clinical search platform built for the National Health Service, from Medwise AI Ltd, registered in England and Wales and now based in Leeds having started in Cambridge. Co founded in 2019 by Dr Keith Tsui, a doctor, and Luis Ulloa, who came from applied machine learning. The distinguishing feature is what it indexes. Alongside national guidance from the National Institute for Health and Care Excellence and other national sources, it ingests an individual trust's own local policies, formularies and antimicrobial guidelines, and answers a natural language question across both in one place. No other vendor in this category holds national and local institutional guidance in the same answer surface: the reference incumbents index published evidence only, and the institutional knowledge tools index only what the hospital wrote. The platform states conformance to the clinical risk management standards that govern health information systems in England, covering both the manufacturer's safety case and the deploying organisation's, and its published terms carry an explicit statement that the platform involves artificial intelligence and automated decision making when returning results and generating responses, framed by the company as something good governance requires it to say plainly. Evidence includes a peer reviewed prospective pilot conducted with a Welsh health board on emergency and acute medicine queries, and an evaluation funded through a national innovation agency which estimated substantial time savings per search. Set against that, a headline claim of a 25 percent consultation time saving was published without a supporting reference, and an independent reviewer testing the platform in general practice reported finding a filtered link list rather than synthesis and no saving of that scale. Both are on the record. Commercial model is unclear from public material: earlier accounts describe free sign up open to any healthcare professional, later third party accounts describe enterprise licensing not available to individual clinicians.

Drug information service delivered as software, from an Atlanta company operating since at least 2014. Co founders Ashish Advani, a pharmacist and the chief executive, and Tulasee Rao Chintha, chief technology officer. The company's own one line description is the clearest statement of what it is: clinical pharmacist reviewed, artificial intelligence generated drug information, so that a health system's own pharmacists spend more time with patients. The architecture is the reason this record matters to the category. A clinician submits a free text clinical question and states how urgently they need it. An engine parses the inquiry; a drug information pharmacist formulates a researchable question from it; the engine generates sub questions using prompt templates and large language models, retrieves against a vector database, and produces a candidate summary; and a board certified drug information pharmacist then reviews that summary, discards it if it does not fit the inquiry, and composes the response that reaches the clinician. Prior pharmacist judgements feed back as reinforcement signal into later retrieval. The assistant is named Sherlock and the vector infrastructure is named openly. Output is a custom literature search with citations rather than a database lookup, spanning guidelines, randomised trials, case reports and tertiary analyses. A second arm serves pharmacy and therapeutics committees with drug monographs, class reviews and formulary management, and that is where the company's stated commercial case sits, since formulary standardisation carries measurable savings. More than 30 health systems are stated to retain what the company calls a virtual drug information pharmacist, and an earlier figure gives more than 10,000 physicians, nurse practitioners and pharmacists using it. The company also runs a one year drug information fellowship that trains the pharmacists who staff the service. Investors include Qlarant Capital and Atlanta Ventures. One published limit belongs on the record because it is unusually direct: the company states plainly that it is not a covered entity or a business associate platform for protected health information and instructs users not to submit any.

Clinical artificial intelligence reference platform from CaryHealth, launched August 2024 on the back of the company's GalenAI acquisition and available at askclair.ai, on web and as native mobile applications. Chief executive Areo Nazari, a pharmacist by training; chief technology officer Matt Hawkins. Record scoped to Clair rather than to CaryHealth, whose other products are a direct to patient platform for manufacturer access programmes, a field medical tool for sales representatives and medical science liaisons, a care gap closure product for health plans, and a nationally licensed pharmacy. The product is organised as six purpose built tools rather than a single answer box: open search, guided clinical summary of a disease or drug, drug interactions, antimicrobial stewardship covering treatment, dose and duration with optional patient demographics, plain language definitions of diagnosis codes, and medication review. Sources named are the Food and Drug Administration drug package inserts, the National Institutes of Health clinical trials database, the National Library of Medicine drug interaction database, clinical guidelines, and any documents the customer uploads. Every response carries links to its references, and the company describes intelligent sourcing that routes a question to the source type it fits. Two things a buyer should read together. The corpus and the six tools are overwhelmingly pharmacologic, so despite the general clinical framing this is closer to a drug and therapeutics reference than to a disease reference of the kind UpToDate or DynaMed sell. And the enterprise offering includes custom document ingestion, white labelling, an application programming interface, and customer defined artificial intelligence guardrails, sold into an industry list that names pharmaceutical manufacturers first. A change worth recording: the August 2024 launch release claimed the semantic search approach produced answers with zero hallucinations. That claim does not appear on the current product material, which uses the more careful language of sourcing exclusively from credible clinical sources with reference links.

Generative clinical reference layer built on ClinicalKey, Elsevier Health's point of care content platform. Launched February 2024 as the publisher's answer to conversational clinical search, and materially expanded in February 2026. Record scoped to ClinicalKey AI rather than to Elsevier, whose portfolio spans scientific publishing and research analytics with no clinical relevance, following the UpToDate and DynaMed precedent. Note also that a separate Elsevier product, ClinicalPath, was assessed for this index in July 2025 and rejected for carrying no artificial intelligence claim; the rejection does not carry over to this record. The corpus is a proprietary, copyright cleared knowledge base refreshed daily, spanning Elsevier point of care content, reference texts including Braunwald's Heart Disease, Goldman-Cecil Medicine and the Nelson Textbook of Pediatrics, society guidelines from bodies including the American College of Cardiology and the European Society for Medical Oncology, and, since February 2026, full text from more than 130 premium journals including The Lancet and the New England Journal of Medicine. Earlier material described third party journals as abstracts only, so the February 2026 release is a real change in what the model can read rather than a restatement. Three mechanisms distinguish it. Responses carry traceability to the specific paragraph in the source rather than to the document. A real time citation validation step checks that a generated response is grounded in the evidence it cites. And the clinician can toggle explicitly between adult and paediatric context, which makes the population a set parameter rather than an assumption. Access is through web, a mobile application with voice dictation, single sign on into the record using the Smart on Fhir standard, and an application programming interface. Continuing education and maintenance of certification credit accrue from use. Available in more than 50 countries. One structural fact belongs on the record: the product was developed in partnership with OpenEvidence, which is separately indexed here and now a direct competitor at scale.

Point of care disease reference from EBSCO Clinical Decisions, a division of EBSCO Information Services, and one of the two reference incumbents this segment is measured against alongside UpToDate. Content rests on a stated seven step evidence based methodology with daily systematic literature surveillance, edited by a named clinical team under editor in chief Dr Peter Oettgen. Record scoped to DynaMed and its generative layer rather than to EBSCO, whose wider portfolio spans library and research databases with no clinical relevance, following the UpToDate and ModMed precedent. The AI is Dyna AI, introduced July 2024 as a retrieval augmented generation capability sold as an add on, extended in February 2026 into Dyna AI Mode, a dedicated interface inside DynaMed, DynaMedex and Dynamic Health that a clinician can toggle in and out of rather than being forced through. The company states the generative layer draws only on its own curated content sets and never on external unvetted sources, labels AI generated output, and links each answer to its origin. Dyna AI Mode also captures questions, responses, references and timestamps in copy ready form and allows a prior query to be re run, which is closer to an audit trail than anything else graded in this category. DynaMedex is a separate joint surface combining DynaMed disease content with Micromedex drug content from Merative, and was named top performing for clinical decision support point of care disease reference in the KLAS 2026 report; DynaMed itself was Best in KLAS in 2021, 2022, 2024 and 2025. A December 2025 announcement extends Dyna AI to generate structured drug monograph and dosing summaries from Micromedex data. EBSCO Clinical Decisions joined the Coalition for Health AI in September 2024 and publishes five principles for the responsible use of AI covering quality, security and patient privacy, transparency, governance and equity. Two scope statements are carried on the record because the company publishes them plainly and they bound what a buyer actually gets: Dyna AI is currently unavailable in the European Union, and DynaMed with Dyna AI for individual subscribers is currently available only in the United States.

Clinical evidence search engine for licensed clinicians, operated by Veracity-Health Inc. and free to verified clinicians and trainees with unlimited searches. The product is built on a retrieval first architecture that the company states is deliberately the reverse of generation first assistants: it searches a corpus given as more than 60 million peer reviewed papers, guidelines and care pathways, ranks what is relevant, applies evidence grading logic the company likens to the work of a guideline methodologist, then generates a short answer with inline citations tied to specific statements. Alongside search it ships drug and interaction checking, a large calculator library, differential and treatment planning views, and specialty tailored literature updates, and it accrues continuing education credit per search, which is the second instance of that model in this category after UpToDate. In March 2026 the American College of Emergency Physicians announced a partnership placing ACEP clinical policies inside the product, confirmed on ACEP's own newsroom and quoted by ACEP president L. Anthony Cirillo. Direction comes from a named international clinical council of sixteen physicians chaired by Scott M. Silvers, formerly chair of emergency medicine at Mayo Clinic and chief medical officer for knowledge management at Optum Health. Founded by Maxime Allouch and Taieb Bennani, both from MIT, and backed by Y Combinator and Gradient Ventures. Two scope notes carried on the record: every published performance figure is vendor run and no peer reviewed evaluation of the product was located, and the company publishes comparison content about other vendors on its own blog, so its characterisations of competitors are a self interested source.

The reference clinical decision support product against which most point of care AI tools position themselves, published by Wolters Kluwer Health and in market for more than thirty years, with content curated by over 7,600 physician authors and editors and reported use by more than three million clinicians across thousands of hospitals worldwide. This record is product scoped to UpToDate and its generative layer rather than to Wolters Kluwer, which also publishes Lexidrug, Medi-Span, Ovid and a large portfolio outside healthcare. UpToDate Expert AI, launched September 2025, is a conversational generative assistant that draws exclusively on UpToDate's expert authored and peer reviewed content rather than the open web. Its distinguishing feature is not the answer but the scaffolding around it. Every response opens with a stated Assumptions section naming what the system took for granted about the patient, for example that the patient is not pregnant, not a child and carries no major comorbidities, and inline Nudges prompt the clinician to supply context the system lacks, such as comorbidities or ICU status. Answers carry inline citations tied to specific statements with single click access to the source topic, a side panel with rationale and supporting material, and drug information drawn from Lexidrug. Clinicians can accrue CME, CE and CPD credit while searching, redeemable in the platform, which the company states is not yet recognised by every accrediting authority. Wolters Kluwer has published an AI principles statement and a validation framework for evaluating clinical AI at the point of care, and named OpenAI as its enterprise model partner in June 2026. Expert AI is included in all United States personal subscriptions and in Pro Plus and trainee subscriptions in Canada, and reaches health systems through UpToDate Enterprise Edition. Chief Medical Officer Peter Bonis MD.

Clinical pathways authoring and governance platform that converts a health system's guidelines, policies and protocols into structured visual decision trees delivered inside the EHR workflow. The product covers clinical pathways and what the company calls ClinApps, guidelines and protocols, policies and regulatory content, antimicrobial stewardship and antibiograms, and embedded clinical calculators, with all content searchable, versioned, governed and measurable. Authoring is visual and requires no code, and pathway logic stays under clinician control without information services involvement after the initial integration, which is the operational claim the company leads on. A community layer lets organisations share and copy each other's pathways across institutional boundaries and across different EHRs. Governance is presented as the core feature rather than a compliance wrapper: named clinical owners, defined review cycles, multidisciplinary review workflows, version history and audit trails, and visible status showing what is live, outdated or under review. The AI scope is deliberately narrow and the company states it plainly on a dedicated page: artificial intelligence is applied to structuring and drafting content from guidelines and PDFs, to supporting updates as evidence changes, and to scaling pathway volume without proportional staffing, under a stated principle of human in the loop at every stage, transparent logic and provenance, and no autonomous clinical decision making. The marketed position is explicitly against opacity, arguing that healthcare needs clarity, transparency and accountability rather than black boxes, and that decision logic should be explicit rather than implicit. Reported at more than one million patient lives impacted, with OSF HealthCare among named implementations and a concentration in paediatric and children's hospital settings. Founded by emergency medicine physicians Eric Leroux and Dan Imler.

Clinical knowledge management platform anchored in anesthesiology and perioperative services, built on the premise that the problem is not missing evidence but the gap between documented protocol and actual practice. Hospitals hold their standards as scattered PDFs, unlisted videos and physical binders, and the company frames the resulting delay in adopting new evidence against the widely cited seventeen year evidence to practice lag. C8 unifies that material into one searchable corpus delivered by role, department and schedule, restructures uploaded documents into mobile navigable formats, and tracks adherence through dashboards for quality leadership. The AI layer is C8 Panda AI, a generative assistant that answers natural language questions sourced exclusively from the institution's own vetted knowledge base, returning inline citations that name both the source document and its author, so a clinician can trace an answer back to the colleague who wrote it. A Knowledge Network lets participating institutions and clinical societies share protocols across organisational boundaries. The platform is explicitly architected to operate without requiring PHI, which is an unusual and materially different privacy posture from most clinical AI. Founded 2019 at Geneva University Hospital by Dr Ido Zamberg, a physician and software engineer who spent fourteen years on enterprise knowledge management at HP and Autodesk, with Dr Olivier Windisch; incorporated in the United States in 2022 with CEO Galia Rosen Schwarz and co founder Tzach Klo. Anesthesia was the first vertical. Reported in use at more than one hundred US hospitals including Mount Sinai, UCSF Health, Brigham and Women's, MetroHealth, UTMB and Dartmouth Health, and named an official collaborator of the American Society of Anesthesiologists. Raised a 12 million dollar Series A in July 2025.

Avo is a clinician built decision support platform whose foundation is content rather than a model: Avo Pathway is a library of digitised medical society guidelines and calculators, which health systems customise or supplement with their own protocols, and which knowledge publishers can embed their intellectual property into. The AI layer sits on top of that. Avo Assistant covers pre charting, care planning, ordering and ambient documentation, and Chart Assist produces a patient snapshot with care gaps, diagnosis and treatment recommendations. Its ambient scribe is therefore not a standalone note writer: it drafts orders, identifies care gaps, suggests dosing, and enhances notes with differential diagnosis, clinical documentation improvement and coding. Two capabilities stand out for this index. It captures conversations with three or more participants across 50 or more languages, naming child, parent, translator and clinician explicitly, which is the hard case most scribes avoid. And its recommendations are anchored to named published guidelines including society pathways, MCG milestones and RAND consensus pathways rather than to model inference alone.

Real world evidence generation, spun out of Stanford's Green Button project in 2020 and built to answer the question a clinician actually has at the bedside when no trial covers their specific patient. GENEVA OS is the underlying operating system across a federated network the company describes as exceeding 300 million anonymized patient records. ChatRWD is a chat based application that generates publication grade observational studies in minutes rather than months, without requiring a data scientist in the loop, and Green Button remains as a consultation service delivering evidence within 48 hours. Alexandria is the company's evidence library of pre computed evidence briefs. Two design decisions distinguish it: ChatRWD is an LLM independent framework, so an institution can maintain its own model security posture, and every answer carries a Real World Fitness Score rating how fit for purpose the underlying dataset was for that specific question, which is an explicit and unusual admission that evidence quality varies by question. Stanford Health Care has moved from using the service to embedding it in physician workflow and clinical notes. The Atropos Evidence Agent is distributed via the Databricks Marketplace under a partnership announced June 2025.

Clinical AI platform built around Almanac Copilot, which the company describes as a Level 1 autonomous EHR agent: it retrieves and summarizes patient data, drafts notes, places orders, and surfaces prioritized alerts, dynamically selecting and chaining tools from a defined set including FHIR functions, a browser, clinical calculators, a Python interpreter, and the EHR database. As a Level 1 agent it acts only on explicit clinician command and requires review and approval of every action. Grounded in peer reviewed evidence in a vector database rather than an unconstrained model, and stated to be free of pharmaceutical advertising, a deliberate contrast with ad supported clinical reference products. On the EHR QA benchmark of 300 common EHR tasks it reported a 74 percent completion rate, matching much larger models. Founded by physician researcher Cyril Zakka MD, whose Stanford work introducing retrieval augmented generation to clinical medicine became one of NEJM AI's most cited papers. Important scope note carried from the source record: the company characterizes the platform as validated through research rather than shipped as a finished product, and it is undergoing clinical validation in academic medical center settings. $10 million seed in April 2026 led by F-Prime with General Catalyst and Lightspeed.

Clinical AI platform pairing ambient scribing with evidence grounded clinical reasoning, which the other scribes do not attempt. Clinicians describe a case in natural language and the platform returns a ranked differential with suggested workups, drafts a problem oriented assessment and plan with inline evidence links, and answers clinical questions with citations drawn from an index of more than 38 million peer reviewed articles, consensus guidelines, and FDA information covering over 154,000 drug compounds. In the Harvard and Stanford ARISE NOHARM evaluation, a physician validated clinical safety benchmark, the platform ranked among the top medical AI systems. The scribing side produces clinic notes, H&P notes, progress notes, discharge summaries, and patient handouts, with assisted EHR workflows across Epic, eClinicalWorks, athenahealth, and Elation. A Developer API exposes the full stack, including clinical question answering, chart summarization, differential generation, care planning, and scribing, so other platforms can embed the clinical intelligence. Independent testing notes reasoning is strongest on common presentations and weaker on complex multi system cases, and the company positions the product as decision support requiring physician judgment on every output. Founded 2021 by Dr Dereck Paul and Graham Ramsey.

AI medical search and clinical decision support used at the point of care, free to verified United States clinicians and monetised through pharmaceutical and medical device advertising, with enterprise deployments embedding the platform inside Epic at named health systems. Two answer modes are offered: Quick Consult returns a short cited answer in seconds, and Deep Consult, introduced in mid 2025, runs a longer agentic synthesis returning a fuller report with more references. Answers are grounded in more than 300 medical journals plus FDA and CDC sources, under official AI partnerships with the New England Journal of Medicine, with JAMA and its eleven specialty journals, with Cochrane Systematic Reviews and with the National Comprehensive Cancer Network for treatment algorithms, alongside content agreements with more than a dozen specialty societies including the American College of Cardiology, the American Academy of Family Physicians and the American College of Emergency Physicians. Ancillary functions include patient handouts, clinical calculators and risk scores, and drafting prior authorisation letters. The company reports more than 757,000 verified clinician users, daily use by over 40 percent of practising United States physicians across more than 10,000 hospitals and medical centres, and approximately 20 million clinical consultations per month as of January 2026. Founded by Daniel Nadler, launched through the Mayo Clinic Platform Accelerate programme, and valued at 12 billion dollars following a Series D in January 2026. Material change in 2026: OpenEvidence terminated access across the European Union and the United Kingdom at the end of April 2026, citing mounting regulatory uncertainty regarding the treatment of AI systems in those markets including the EU Artificial Intelligence Act, so the product is now United States centred with primary verification built around the US National Provider Identifier. Unusually for this index, the platform has been the subject of a peer reviewed systematic review of eleven independent evaluations, which found it avoids fabricated citations and performs best on structured guideline based questions while accuracy varies on complex cases.