Sustain
Insuring AI-powered
health and wellness
AI is changing how care is delivered, and it is creating risks that today’s medical malpractice, product liability and technology E&O policies were never written for. Sustain identifies, evaluates and places coverage for those risks.
The problem
AI is changing the way health and wellness care is delivered
Typical health and wellness coverage is built around the negligence, errors and omissions of licensed professionals. It is usually silent on AI-enabled tools, creating uncertainty in coverage and leaving clinicians, delivery organizations and developers unsure where liability sits.
Sustain was built to close that gap: an analytical platform for identifying and evaluating AI risk, carefully crafted policy terms and forms, coverage for harms such as bodily injury, and strong relationships with carriers and clients.
Where the risk is
Three kinds of AI, three kinds of exposure
We organize health-care AI by how it enters the care pathway. Each category carries a different liability profile, and each calls for coverage designed around it. Open a category for the detail.
Clinical AI
AI-enabled medical devices and software used for detection, diagnosis and triage, principally in radiology and digital pathology.
- Over 1,000 FDA-authorized AI-enabled devices
- On-label and off-label clinician use
- Laboratory-developed tests and ambulatory diagnostics
Infrastructure AI
AI embedded in clinical workflow: ambient scribes, report writing, predictive care, automated measurement and data capture.
- Workflow and documentation errors
- Indemnification terms that leave gaps
- Organization-level governance exposure
Interactive AI
Chatbot, telemedicine and remote-device care: outreach, triage, monitoring and therapeutic engagement, sometimes without human supervision.
- Evolving standards of care
- Clinician-centered liability frameworks that may not fit
- Emerging litigation against generative-AI providers
How liability is covered
Coverage for every party in the loop
A single AI-related error can touch the patient, the clinician who relied on the tool, the organization that deployed it and the developer that built it. Carriers with policy exposure to any of these parties may have unintended coverage obligations. Sustain designs coverage for some or all of the following parties.
-
Patients
Protection of the patient through covered claims when algorithmic advice or an AI medical device causes harm.
-
Clinicians and delivery organizations
Protection from liability and defense costs when AI tools are used in care, alongside existing malpractice coverage rather than in conflict with it.
-
AI developers and providers
Coverage for negligence, error or defect in AI-based products and services.
How we work
From risk assessment to bound coverage
Sustain works with health-care delivery organizations, wellness providers, clinicians, carriers and developers.
Identify
We identify and assess the needs of carriers and insureds and the specific risks of the AI in use: the model(s), its deployment, its data and the clinicians around it.
Evaluate
Risks are evaluated through proprietary qualitative and quantitative models, drawing on our inventory of parameter sets for specific devices, services and applications, or bespoke assessment of novel technologies.
Design
Policy forms are customized to the risks actually borne by the carrier and the client, and written to work with other coverages rather than against them.
Place
We streamline placement with highly rated carriers and stay involved as AI, regulation, and the standard of care continue to move.
Underwriting platform
Models that work from the error up
Sustain’s platform models the claims risk process from the ground up: risk assessment calibrated to well-understood claims history and AI error profiles.
Medical and Wellness Risk module 
Claims module 
Loss Allocation module Resources and news
Briefings and news coming soon
Short briefings for carriers, risk managers, clinicians and developers are in preparation. Check back soon, or contact us if you would like to be notified when they are available.
About Sustain
Built around one problem
Sustain focuses on risk management for health-care and wellness-related applications of AI. We understand the interplay between AI, medical practitioners, patients, and professional and product liability, and we bring expertise across AI, insurance, quantitative methods, and insurance law and regulation.
Sustain was founded by a team of senior professionals in insurance, risk management and insurance law with a track record of providing insurance and other solutions to novel risks.
Disclaimer
Sustain Insurance Services, LLC is a licensed surplus lines insurance broker. Nothing on this website is an offer of insurance, and coverage cannot be bound through it; policies issued by surplus lines insurers are generally not protected by state guaranty funds.
Read the full disclaimer
Sustain Insurance Services, LLC (“Sustain”) is a licensed surplus lines insurance broker. Coverage placed by Sustain may be written by surplus lines (non-admitted) insurers. Surplus lines insurers are not subject to the financial solvency regulation and enforcement that apply to admitted insurers, and in the event of insolvency, policies issued by them are generally not protected by state insurance guaranty funds. Licensing details are available on request.
Nothing on this website is an offer to sell insurance, a quotation, or a representation that coverage is available. Insurance coverage cannot be bound, changed, renewed or cancelled through this website, by e-mail, voice mail or any other electronic submission. No binder, policy, endorsement or change to coverage is effective unless and until it is confirmed in writing by a licensed representative of Sustain or the issuing insurer.
Descriptions of policies and coverages on this website are general summaries provided for informational purposes only. Actual coverage, terms, conditions, exclusions and limits are governed solely by the policy as issued by the insurer. Sustain makes no warranty as to the accuracy or completeness of the information on this website, which may change without notice.
Information on this website does not constitute legal, tax, actuarial, medical or other professional advice. If you have questions in those areas, please consult a qualified professional.
Links to third-party websites and vendors are provided for convenience only. Sustain does not control, endorse or accept responsibility for their content.
Use of this website and reliance on anything in it is at your own risk. To the fullest extent permitted by law, Sustain will not be liable for any loss or damage arising out of or in connection with the use of this website or its temporary unavailability, including indirect or consequential loss and loss of data or profits.
This website uses cookieless, aggregated analytics (Cloudflare Web Analytics) to count visits. No cookies are set and no personal data is collected.
The FDA has authorized more than a thousand AI-enabled medical devices, including software, for disease detection, diagnosis and triage. Most have come through the 510(k) clearance pathway, principally in radiology and pathology, and others operate under approaches such as laboratory-developed-test enforcement discretion that may carry few FDA-related obligations.
Clinicians may use approved tools on-label or off-label. Governance, regulatory and legal liability frameworks for AI-based decision making are still evolving, particularly where clinicians have limited transparency into the algorithms they rely on.
To date the liability framework has focused on clinician users and health-care organizations, especially where AI developers disclaim liability contractually. Where AI is used to increase the number of cases over which clinicians must make professional decisions, the overall liability borne by the clinician can rise.
Who this affects
- Hospitals and clinics
- Radiology and pathology groups
- Diagnostic laboratories
- Ambulatory and lab-centered services
- Developers of diagnostic algorithms
Health-care delivery organizations are increasingly using AI in clinical workflow: ambient scribes, clinical report writing, diagnostic review, reasoning and prediction. The emphasis is on efficiency and improved patient outcomes, and the tools often sit between the clinician and the record.
The allocation of risk between organizations and AI developers is often left unclear. As with clinical AI, governance and liability frameworks are evolving, especially when clinicians and other professionals may have limited insight into how a given AI-based tool behaves.
The practical result is that organizations may carry exposure they did not intend to carry, and that existing policies do not clearly address.
Who this affects
- Hospitals and health systems
- Physician groups adopting ambient documentation
- Wellness providers using predictive tools
- Workflow and documentation vendors
Providers are increasingly turning to chatbot-based care delivery for patient outreach, management, triage, monitoring, decision support and therapeutic engagement. These tools have shown promise in delivering care efficiently at scale, both stand-alone and alongside licensed professionals.
Governance, regulatory and legal frameworks for this kind of care are in flux. Standards of care are still forming, and existing liability frameworks are built around clinicians, so they may not reflect the actual sources of risk or the nature of the potential harms.
Generative-AI providers have already been sued for injuries to users, including self-harm, and have settled with plaintiffs. The exposure is real today, not hypothetical.
Who this affects
- Digital-health and telemedicine platforms
- Behavioral-health and wellness apps
- Remote monitoring services
- Developers of conversational agents
Briefings on AI liability in care
Short briefings for carriers, risk managers, clinicians and developers are in preparation. Check back soon, or contact us if you would like to be notified when they are available.
News and insights
News and insights on AI liability in health care are in preparation. Check back soon, or contact us if you would like to be notified when they are available.
This document is in preparation
We are finishing this item. Contact us if you would like a copy as soon as it is available.
Contact
Tell us about the AI you use
Whether you are a carrier looking at a new line, a delivery organization reviewing your exposure, or a developer being asked for proof of coverage, a short conversation is the right first step.
Thank you.
We have your message and will respond shortly.