Can AI Really Write Your Clinical Evaluation Report?

Small Business, Smart Start: AI Without Complex Infrastructure Artificial Intelligence is creating new opportunities for businesses to improve productivity, simplify workflows, and reduce repetitive administrative work. However, for many small medical device companies, adopting AI can feel overwhelming. Where should you begin? Which processes are worth automating? Do you need expensive infrastructure or a dedicated development team? And how can you ensure that an AI assistant delivers reliable results? In this episode of Medical Device Made Easy, we welcome Helder Conceicao to explore how small businesses can take their first practical steps with AI agents. With experience in service engineering, business processes, and supporting AI adoption in small companies, Helder brings a practical perspective to a topic that is often presented as more complicated than it needs to be. Understanding AI Agents: Beyond Traditional Chatbots Many professionals have already experimented with AI tools to answer questions, draft emails, or summarize information. But an AI agent introduces a different way of thinking about how AI can support business processes. Rather than focusing only on individual conversations, the discussion explores how AI assistants can support defined tasks and workflows. Helder explains the differences between traditional chatbots and AI agents, including what these solutions can accomplish and where human involvement remains necessary. The objective is to help business owners understand the practical possibilities before investing in a solution. Choosing the Right First AI Project One of the biggest challenges in AI adoption is deciding where to begin. Businesses may have dozens of processes that could potentially benefit from automation, but not every task is a good starting point. The conversation explores how to identify a first project based on a genuine operational need. For medical device companies, potential areas to investigate include: Documentation: Supporting document preparation, organization, and information retrieval. Quality and Regulatory Affairs: Assisting with administrative activities and information management while maintaining appropriate human review. Field Service: Exploring ways to support service-related workflows and routine processes. The key is to identify a manageable use case with a clear purpose and measurable value. From an Idea to a Working AI Assistant Once a company has identified a suitable process, what happens next? The episode explores the journey from an initial business problem to a functioning AI solution. This includes understanding what information the company needs to prepare, how the workflow should operate, and what technical resources may be required. An important part of the discussion is determining whether an existing AI platform can meet the company's needs or whether additional development and infrastructure are necessary. For smaller organizations, these decisions can have a significant impact on project complexity, cost, and implementation time. How Much Does AI Implementation Really Require? Cost and resources are common concerns for businesses considering AI. Does every AI project require a large budget? How much time should the company expect to invest? And how involved does the internal team need to be? Helder discusses the practical considerations behind planning an AI implementation, including the importance of defining the project's scope and understanding the resources required. The goal is to help businesses approach AI adoption with realistic expectations rather than assuming that every solution requires a major technology investment. Keeping Humans in Control For medical device companies, automation must be considered alongside reliability, accountability, and appropriate oversight. Not every task should be delegated to AI without review. The conversation addresses how to determine which activities an AI assistant can support independently and which requi

October brings another important round of regulatory developments for the medical device and IVD industry. From updated IVD classification guidance in Europe and potential changes to the UK medical device licensing framework, to new FDA expectations for robotically assisted surgical devices, radiology AI, PMA submissions, and international regulatory reliance, manufacturers have several developments worth following closely. Here is our Medical Device Regulatory Update for October 2026, supported by regulatory intelligence from MedBoard. 🇪🇺 European Union – Revised IVD Classification Guidance The European Commission has published MDCG 2020-16 Rev.5, updating the guidance on classification rules for in vitro diagnostic medical devices under the IVDR. For IVD manufacturers, classification is a fundamental regulatory decision that can affect the conformity assessment route and regulatory strategy. 👉 MDCG 2020-16 Rev.5 – Guidance on Classification Rules for IVDs 🇬🇧 UK – The Door Opens to Direct MHRA Licensing Proposed amendments to the UK Government's Health Bill could create powers supporting the future development of a medical device licensing regime. The amendments also address information-sharing powers for medicines and medical devices and measures intended to make medicines and medical-device legislation easier to update. This is an important development to watch for manufacturers planning access to the UK market. 👉 MHRA Regulatory Reform Amendments – Health Bill 🇬🇧 MHRA Warning – Incompatible Components Continue to Cause Serious Incidents The MHRA has highlighted ongoing safety concerns involving patient hoists and incompatible components. The warning is an important reminder that medical-device safety does not stop with individual component compliance. Compatibility between components, accessories and systems can directly affect patient safety. Manufacturers and healthcare organizations should pay close attention to compatibility information, instructions for use and the way devices are combined in real-world environments. 👉 MHRA – Steps to Prevent Avoidable Harm from Patient Hoists 🇨🇭 swissdamed – Importer Linking Is Now Operational Another important development comes from Switzerland. Swiss importers can now link their CHRN to imported devices in swissdamed. This adds another operational consideration for manufacturers and importers working in the Swiss medical-device market, particularly as swissdamed becomes increasingly important within the country's regulatory infrastructure. 👉 Swissmedic – Information for Importers in swissdamed 🌍 Rest of the World 🇺🇸 FDA – Robotically Assisted Surgical Devices The FDA has issued draft guidance addressing robotically assisted surgical devices. The guidance covers areas including non-clinical testing, clinical evidence and labeling for teleoperated surgical systems. For manufacturers developing robotic surgical technologies, understanding these expectations early can help inform development, testing and premarket strategy. 👉 FDA – Robotically Assisted Surgical Devices: Premarket Submissions 🤖 FDA Rejects Proposed 510(k) Exemption for Radiology AI Devices AI remains firmly on the regulatory agenda. The FDA has rejected a proposed exemption from premarket notification for certain radiology computer-aided detection, diagnosis and triage devices. The practical message for manufacturers is clear: these products must continue to obtain the applicable FDA clearance before being marketed. 👉 FDA – Radiology Computer-Aided Detection, Diagnosis and Triage Devices 🇺🇸 FDA Moves PMA Applications Toward Structured Electronic Submissions The FDA is also moving toward a more structured approach to electronic Premarket Approval (PMA) applications. Draft guidance introduces electronic resources and templates intended to improve submission consistency and review efficiency. Manufacturers working on PMA submissions should follow this development as electronic submission structures continue to evolve. 👉 FD

Artificial Intelligence may already be part of your Quality Management System, even if your organization has never formally implemented an AI tool. Regulatory Affairs professionals may use ChatGPT to research requirements. Quality teams may use AI to help draft procedures or summarize meetings. Engineers may ask AI to brainstorm potential failure modes. Post-Market Surveillance teams may experiment with AI to analyze complaints. This creates an important challenge: AI may already be inside your QMS—even though your QMS doesn't know it. Markdown collé The answer isn't necessarily to ban AI. The answer is to understand where it is being used and introduce controls proportionate to the risk. The Shadow AI Problem Employees don't always wait for management to formally implement AI. When a tool can turn a three-hour activity into a thirty-minute task, people naturally become interested. The Quality challenge isn't simply that AI is being used. The problem is that the organization may not know: what activities AI supports, what information employees provide to it, how outputs are verified, and how heavily people rely on those outputs. This is what the episode describes as Shadow AI. Markdown collé Not Every AI Use Has the Same Risk Asking AI to improve the wording of an email isn't equivalent to asking it to determine whether complaint data indicates a safety signal. The key question is: What happens if the AI is wrong? An error in a rewritten email may have limited consequences. An incorrect conclusion about complaint reportability or patient safety could have significantly different consequences. AI governance should therefore be based on intended use and risk, rather than treating every AI application identically. Markdown collé Green, Amber, Red: A Practical AI Framework The episode introduces a simple traffic-light approach. 🟢 Green — Assist AI improves productivity without materially driving a regulated decision. Examples include improving English, formatting documents, brainstorming, summarizing meeting notes or helping structure a presentation. Normal human review, together with appropriate confidentiality and information-security controls, may be sufficient for these lower-risk applications. 🟠 Amber — Verify AI begins influencing regulated work. Examples include helping draft an SOP, suggesting potential FMEA failure modes, summarizing complaints, researching regulatory requirements or suggesting possible CAPA root causes. The principle here is simple: AI can propose. A qualified person must verify. AI can help locate information, but it shouldn't automatically become the regulatory evidence supporting a decision. 🔴 Red — Assess Before Deployment Here, AI makes or materially drives a critical decision. Examples could include automatically determining complaint reportability, accepting suppliers, closing CAPAs, deciding risk acceptability or generating clinical conclusions without independent qualified evaluation. These uses call for formal assessment before the organization relies on the system. Markdown collé Human Review Is Not Automatically Verification “We have a human reviewing everything” sounds reassuring. But what does review actually mean? Generative AI can produce incorrect information that looks professional and convincing. Reading an AI-generated regulatory assessment and thinking “looks good” isn't necessarily meaningful verification. The episode proposes another important principle: Trust the source, not the sentence. Regulatory requirements should be checked against the regulation. Complaint-analysis results should be checked against the underlying data and methodology. Technical suggestions should be evaluated by qualified personnel. Markdown collé Five Steps for Bringing AI Under Control Rather than immediately writing an AI SOP, the episode recommends starting with discovery. First, inventory where AI is already being used. Then classify each use according to what could happen if the output is wrong. Next,

Developing complex Software as a Medical Device (SaMD) can quickly become challenging.

Quality and Regulatory professionals often find themselves in a frustrating position.