Doximity's AI Dosing Errors Expose Healthcare Marketing's Clinical Liability as Systems Race to Deploy Patient Tools

Popular artificial intelligence models are calculating medication doses incorrectly, with some errors reaching dangerous levels, according to research released by Doximity in 2026 . The findings arrive as healthcare systems race to deploy AI-powered patient engagement tools, symptom checkers, and medication management platforms, creating a collision between marketing promises and clinical safety that could expose organizations to regulatory scrutiny and erode the patient trust that digital health strategies depend on.

The timing compounds the risk. Enrollment periods for Medicare drug plans are seeing premium increases of 50% in 2027, from $34 to $51 monthly on average, driving beneficiaries to seek online tools and AI chatbots for medication cost comparisons and dosing information . When patients turn to digital channels for drug guidance during periods of financial stress, the accuracy of AI-generated medical information becomes a market differentiator and a legal exposure point simultaneously.

Healthcare organizations that have marketed AI tools as patient convenience features now face a harder question: How do you promote digital engagement when the underlying technology produces clinical errors? The gap between marketing messaging about "AI-powered care" and the documented reality of dosing mistakes creates vulnerability on multiple fronts, malpractice risk, FTC advertising scrutiny, and the reputational damage that follows when patients discover the tools don't work as advertised.

This matters beyond medication management platforms. The Doximity findings signal a broader reliability problem across consumer-facing AI health tools. If large language models miscalculate drug doses, a relatively structured clinical task with clear protocols, their performance on more ambiguous patient inquiries (symptom assessment, care navigation, treatment education) becomes suspect. For healthcare marketers who have built patient acquisition strategies around AI chatbots, telehealth triage systems, and automated care coordinators, the liability question is no longer theoretical.

The Marketing-Clinical Gap Widens as AI Adoption Accelerates

Healthcare systems launched hundreds of patient-facing AI tools between 2024 and 2026, many marketed as ways to improve access and reduce call center volume. Marketing departments promoted these platforms in digital advertising, website redesigns, and patient portal campaigns. The value propositions were consistent: 24/7 availability, instant answers, personalized guidance.

The Doximity research reveals that consistency in marketing did not translate to consistency in clinical accuracy . Medication dosing requires precise calculations based on patient weight, renal function, drug interactions, and condition-specific protocols. AI models trained on general medical literature can produce plausible-sounding but clinically incorrect recommendations, a phenomenon researchers call "hallucination" but patients experience as harm.

The financial pressure on Medicare beneficiaries amplifies the risk. With drug plan premiums jumping 50% in 2027, seniors are shopping aggressively for cost savings and turning to online resources to compare formularies and understand dosing options . Robert Travis Scott of Abingdon, Virginia, saved $1,000 annually by switching from a $121.90 monthly AARP-UnitedHealthcare plan to a $19.30 Humana plan with similar benefits . As patients like Scott conduct more of their medication research online, the AI tools they encounter (whether on insurer websites, health system portals, or pharmacy platforms) become de facto clinical advisors.

Marketing leaders at health plans and provider organizations built patient engagement strategies assuming AI tools would enhance the consumer experience. The dosing error findings force a recalibration. Tools marketed for convenience may create clinical incidents. Platforms promoted to reduce costs may generate liability expenses that dwarf any operational savings.

Regulatory Crosshairs: FTC, FDA, and State Attorneys General

The regulatory environment for AI health tools tightened considerably in late 2026. The FDA's launch of the Neuroscience Advancement and Innovation Nexus (NAIN) in October 2026 signals heightened agency attention to the clinical validity of medical products, a category that increasingly includes software and AI systems . Acting FDA Commissioner Kyle Diamantas stated the initiative reflects "the Trump Administration's drive to transform health care innovation" while ensuring "regulatory science and cross-center collaboration keep pace with rapid advances" .

NAIN coordinates oversight across the FDA's drug, biologics, and device centers . This cross-center approach matters for healthcare marketers because many AI tools occupy regulatory gray zones, marketed as patient education or administrative support but functioning as de facto clinical decision aids. When an AI chatbot recommends a medication dose or suggests treatment alternatives, it crosses from marketing tool to medical product.

The FTC adds another enforcement layer. The agency has increased scrutiny of healthcare advertising claims, particularly around AI capabilities. Marketing materials that promise "clinically validated" AI or "physician-level accuracy" without supporting evidence invite enforcement action. After the Doximity revelations, any healthcare organization marketing AI tools for medication guidance faces heightened risk that promotional claims will be tested against actual performance data.

State attorneys general represent a third enforcement vector. Several states have launched investigations into healthcare AI tools following patient complaints. Marketing departments that approved language claiming AI systems provide "safe" or "reliable" medical guidance may find those statements used as evidence in consumer protection cases if the systems produce errors.

The Patient Trust Equation: Short-Term Gains, Long-Term Erosion

Healthcare marketing has shifted toward digital-first patient acquisition over the past three years. Organizations invested in SEO, paid search, and AI-powered website experiences to capture patients researching symptoms and treatment options online. The strategy worked, digital channels now drive 40-60% of new patient volume at many systems.

The Doximity findings threaten that acquisition engine. Patients who receive incorrect medication information from an AI tool don't just abandon that tool: they lose confidence in the organization behind it. A health system that promotes an AI chatbot as a patient convenience feature, only to have that chatbot provide dangerous dosing advice, faces reputational damage that extends beyond the specific incident.

The trust problem compounds in value-based care arrangements. Health plans and provider organizations increasingly earn revenue by keeping patients healthy and managing chronic conditions effectively. Medication adherence is central to those outcomes. If patients stop trusting the digital medication management tools their health plan or doctor's office provides, because those tools have been shown to produce errors, adherence suffers, outcomes worsen, and value-based payments decline.

Marketing leaders focused on patient lifetime value need to calculate the cost of trust erosion. Acquiring a new primary care patient costs $200-$500 depending on market and specialty. Retaining that patient over five or ten years generates substantially higher revenue than the acquisition cost. An AI medication error that drives a patient to a competitor destroys acquisition investment and future revenue simultaneously.

The Medicare premium increases create an immediate test case . Beneficiaries shopping for drug plans during open enrollment (which runs through December 7, 2026) will use online tools to compare options . Health plans that market AI-powered plan comparison tools or medication cost calculators must ensure those tools produce accurate recommendations. A beneficiary who selects the wrong plan based on faulty AI guidance will blame the plan, not the algorithm, and will switch at the next enrollment period.

Recalibrating Your AI Marketing Strategy: From Promise to Proof

Healthcare organizations cannot abandon AI patient tools, the operational benefits and patient expectations are too strong. But marketing strategies must shift from promoting AI capabilities to demonstrating AI reliability. This requires three changes:

First, audit every patient-facing AI tool for clinical accuracy before marketing it. Doximity's research shows dosing errors occur across popular AI models, which means assumptions about vendor-supplied technology cannot be trusted. Organizations need independent clinical validation, ideally by specialty-appropriate physicians, before marketing any AI tool that touches medication management, symptom assessment, or care recommendations.

Second, revise marketing language to reflect AI limitations. Instead of positioning AI tools as replacements for clinical judgment, frame them as preliminary screening or information-gathering resources that connect patients to human expertise. This positioning reduces liability exposure and sets appropriate patient expectations. Marketing copy that acknowledges what AI cannot do builds more trust than copy that overstates what it can.

Third, implement monitoring and error reporting systems before patients discover problems. Marketing campaigns that drive traffic to AI tools should be paired with clinical quality monitoring that tracks error rates, patient complaints, and near-miss incidents. Discovering an AI medication error through internal monitoring allows for controlled remediation. Discovering it through a patient harm event triggers lawsuits, regulatory investigations, and media coverage.

The FDA's NAIN initiative provides a regulatory roadmap . The program's focus on coordinating oversight across drug, biologics, and device centers signals that regulators view AI health tools as medical products requiring evidence of safety and effectiveness . Marketing strategies that anticipate this regulatory framework, by establishing clinical validation processes and limiting promotional claims to proven capabilities, position organizations ahead of enforcement rather than reacting to it.

The 1ness Take

The Doximity AI dosing error findings mark an inflection point for healthcare digital marketing. Organizations face a choice: continue marketing AI tools based on potential benefits while hoping clinical errors remain isolated incidents, or rebuild AI marketing strategies on a foundation of validated performance and transparent limitations.

We recommend the latter, and the approach requires marketing leaders to drive the change rather than wait for clinical teams to impose restrictions. Marketing departments should initiate AI tool audits and performance validation, not because compliance requires it yet, but because patient trust and liability exposure demand it. The organizations that move first gain competitive advantage: they can market AI capabilities with clinical evidence backing their claims while competitors retreat after patient incidents expose overstated promises.

Practically, this means three immediate actions for healthcare marketing teams:

Conduct a marketing-clinical alignment sprint. Convene marketing, clinical leadership, legal, and IT for a 30-day review of every patient-facing AI tool currently promoted through marketing channels. Map marketing claims against clinical validation evidence. Where gaps exist, and Doximity's research suggests they will, either obtain validation or revise marketing language. This sprint prevents the scenario where marketing promises exceed clinical reality.

Develop tiered AI marketing guidelines based on clinical risk. Not all AI tools carry equal liability exposure. An AI-powered appointment scheduler that suggests available time slots poses minimal clinical risk. An AI chatbot that recommends medication changes or interprets symptoms poses substantial risk. Create internal marketing standards that require higher levels of clinical validation and legal review for high-risk AI applications, with approved language libraries that marketing teams can deploy without creating new liability with every campaign.

Build patient transparency into AI marketing as a differentiator. Rather than viewing AI limitations as marketing obstacles, position transparency about what your AI tools can and cannot do as a trust-building strategy. Marketing language that acknowledges "This tool provides preliminary information, your care team makes final decisions" differentiates your organization from competitors still marketing AI as a clinical replacement. In a market where AI errors are now documented, honesty about AI limitations becomes a competitive advantage.

The Medicare premium crisis illustrates the opportunity . With beneficiaries facing 50% premium increases and actively shopping for plans, health insurers that market medication cost calculators and plan comparison tools have immediate patient attention. The organizations whose tools provide accurate, validated recommendations will capture enrollment. Those whose tools produce errors will face both lost enrollment and regulatory scrutiny. Marketing accuracy becomes market share.

Looking ahead, FDA oversight through NAIN will likely expand to cover AI health software more comprehensively . Marketing organizations that establish clinical validation processes now will adapt smoothly to future regulation. Those that continue marketing unvalidated AI tools will face retrospective compliance requirements, potential enforcement actions, and the need to pull marketing campaigns mid-flight when tools fail validation.

The strategic insight: healthcare marketing built on AI promises is fragile; marketing built on AI proof is defensible. In a regulatory environment that increasingly treats healthcare AI as medical products requiring evidence, and a patient environment where AI errors are documented and publicized, marketing strategies must shift from innovation storytelling to performance demonstration. The organizations that make this shift in early 2027, ahead of both competitive pressure and regulatory mandates, will own the market position as trusted AI healthcare providers while competitors manage patient incidents and legal exposure.

The Takeaway

• Audit before you market. Review every patient-facing AI tool for clinical accuracy before promoting it in marketing campaigns, with particular scrutiny on medication dosing, symptom assessment, and treatment recommendation features.

• Shift marketing language from capability to validation. Replace general AI benefit claims ("24/7 intelligent care guidance") with specific, evidence-backed statements ("Appointment scheduling tool with 99.2% accuracy rate validated by independent clinical review").

• Establish cross-functional AI governance that includes marketing. Marketing leaders should co-chair AI oversight committees with clinical and compliance executives, ensuring promotional strategies align with clinical validation evidence and regulatory requirements from campaign inception rather than review.

The AI medication dosing errors Doximity documented in 2026 are not isolated technical failures: they are symptoms of healthcare organizations deploying and marketing technology faster than they can validate it . Marketing departments accelerated that deployment by creating patient demand for AI tools through digital campaigns. Now marketing must lead the quality correction by ensuring the tools they promote actually work as advertised. The Medicare enrollment period running through December 7, 2026, offers an immediate test : organizations whose marketed AI tools help beneficiaries navigate 50% premium increases with accurate information will gain market share and trust, while those whose tools produce errors will lose both.