Beena Wood

Glowing neural network inside a transparent antibiotic capsule surrounded by a large language model on a black background. Illustration of the concept of AI assistance in pharmacy and drug discovery | Image Credit: ©Dragon Claws -stock.adobe.com

For the past 2 years, the conversation about artificial intelligence (AI) in pharmacovigilance (PV) has largely revolved around the question, can the system explain itself? Can it show its reasoning, cite its sources, produce an audit trail? These are genuinely important qualities. But 2026 has been the year the industry, and its regulators, started asking a harder question: not what the system can explain after the fact, but what it prevents from happening in the first place, and how that is governed over time.

That shift matters most in post-marketing changes: the ongoing stream of signal detection, safety labeling updates, and case-processing decisions that don’t stop once a product reaches market. This is where AI’s promise is most real, and where the cost of getting governance wrong is most direct.

An aircraft’s black box and its altitude alarm serve different purposes. The recorder lets investigators reconstruct what happened after an incident, valuable, but too late for anyone on board at the time. The altitude alarm intervenes before a threshold is crossed, while there is still a chance to change the outcome. Most of the AI governance conversation in PV has been built around the recorder. Comparatively, little attention has gone to what it would take to build-in the ability to intervene before a missed or delayed signal becomes a patient safety issue.

Two enforcement actions illustrate exactly what that gap looks like in practice. In March 2026, an FDA warning letter to Novo Nordisk

 documented case-handling failures, including a report excluded from mandatory reporting on assumed causality grounds, and a patient death case invalidated for a missing identifier that the FDA’s own inspectors later found present in the company’s source documents. Notably, the system involved wasn’t AI-based at all. A conventional, rule-based process failed because nothing was actually enforcing the rule, and no one was monitoring closely enough to catch it.

In April 2026, the FDA issued its first warning letter to explicitly name AI overreliance as an inspection finding,

 this time against a drug manufacturer that had used AI agents to generate compliance documents without human review. When investigators found the company hadn’t conducted required process validation, the response was that the AI agent had never flagged it as necessary.

Read together, these 2 letters trace the full arc of governance failure in post-marketing change management: rules that exist on paper but aren’t enforced, and accountability that gets quietly delegated to a system incapable of holding it.

Regulators Are Converging Faster than the Market Has Noticed

This isn’t a scattered set of national concerns. Three threads illustrate just how aligned regulators now are, despite different mandates and methods.

First, none of them believe a single point-in-time approval is sufficient anymore. The Council for International Organizations of Medical Sciences (CIOMS) Working Group XIV report on AI

 in pharmacovigilance, the product of 3 years of work across regulators, industry, and academia, sets out 7 core principles for AI in PV, explicit that explainability is one principle among 7, not the standard on its own, and that human oversight must be risk-calibrated and architecturally enforced.

The European Medicines Agency’s (EMA’s) Network Data Steering Group has translated that same direction into dated commitments

: new AI-enabled signal detection capabilities for authorized products rolled out in the second quarter of 2026, with AI-based suspected unexpected serious adverse reactions (SUSAR) screening for investigational products following in 2027, alongside a life cycle management principle in the 

EMA-FDA Guiding Principles of Good AI Practice

 published in January 2026, requiring scheduled monitoring and periodic re-evaluation to catch data drift throughout an AI system’s life, not just at launch.

The United Kingdom’s National Commission into the Regulation of AI in Healthcare reached the identical conclusion from a public-facing consultation of more than 700 respondents:

 between 61% and 68% of every stakeholder group, patients, clinicians, providers, and industry alike, said current post-market surveillance requirements are not sufficient for how AI systems actually behave once deployed.

Second, all 3 sources agree that oversight has to scale with risk and context rather than apply uniformly. The EMA-FDA principles pair a risk-based approach with a requirement for a clearly defined context of use, on the basis that risk cannot be assessed in the abstract, only against a bounded purpose. The UK Commission’s evidence shows broad support for that same logic, while also surfacing where consensus breaks down: 77% of patients and the public described the current regulatory framework as too loose, while 65% of industry respondents described it as too restrictive. Everyone wants proportionality. Nobody agrees yet on which direction the dial needs to move, which is exactly the kind of ambiguity PV teams should expect to navigate rather than wait out.

Third, transparency and human oversight are treated as the mechanism for accountability, not a nice-to-have. The EMA-FDA principles call for plain-language disclosure of an AI system’s context of use, performance, and limitations, addressed to the intended audience rather than buried in technical documentation. The UK Commission’s evidence found the same expectation from the public side: 77% to 88% of respondents across every group said the current liability framework has significant gaps, tied directly to how difficult it is to trace accountability through an AI system whose reasoning isn’t transparent. Regulators are drawing a straight line from explainability to who is answerable when something goes wrong.

The FDA and EMA have also formalized a standing cluster on AI in pharmacovigilance,

 meeting every 2 to 3 months with Japan’s Pharmaceuticals and Medical Devices Agency and Health Canada as observers, working through use cases spanning the full PV workflow from adverse event extraction through signal evaluation. Regulators are deploying AI in the same workflows they will be inspecting industry for. As an MHRA spokesperson put it earlier this year, the agency’s consistent position is that AI should augment expert judgement, not replace it.

The Risk Explainability Quietly Makes it Worse

There’s a subtler problem worth naming directly. When AI produces fluent, well-explained output, human reviewers tend to scrutinize it less, not more. A confident, well-formatted recommendation creates an illusion of verification: the reviewer feels they’ve checked something they’ve actually just approved. In a PV environment where a case manager may be processing hundreds of cases a day with AI assistance, that’s a foreseeable consequence of deploying explainable AI without controls on reviewer behavior, not a hypothetical one. Complete governance has to monitor what happens to human judgment when the system is working well: approval velocity, override rates, whether a required review step actually happened, not just whether it was scheduled.

What This Means for How Post-Marketing Changes Actually Get Made

Put these threads together and a practical picture emerges. Effective AI governance for post-marketing change management treats deterministic and probabilistic processing as a deliberate choice, not a default. Decisions that need the same correct answer every time, whether a report meets a reporting threshold regardless of assumed causality, for instance, are handled by fixed logic that cannot be configured around. Judgment calls that genuinely require interpretation are handled by AI reasoning, scoped tightly to the question at hand, with confidence thresholds that trigger escalation rather than confident silence when uncertainty is high. Governance itself has to be evidenced across the full lifecycle, design, validation, and live monitoring, not documented once and assumed to hold.

This is consistent with conversations the author has had with customers and echoes a pattern seen repeatedly the author’s experience at a large biopharma safety organization. The appetite has shifted noticeably: less patience for AI framed purely as incremental case-processing efficiency, more urgency around faster, more defensible signal detection and inspection-readiness that’s built in rather than retrofitted. Several enterprise sponsors have independently raised interest in standardizing signal detection methodology across the industry, precisely so regulatory acceptance doesn’t have to be re-litigated company by company. That’s where the practitioners doing this work every day are already heading.

None of this means explainability is unimportant. It’s necessary. But it answers only 1 part of the question that post-marketing change management now demands. The more useful measure of readiness isn’t whether a system can explain itself when asked. It’s whether anyone would have noticed if it were no longer able to.

Warning Letter to Novo Nordisk Inc., MARCS-CMS 717576 / FDA Ref. 26-HFD-45-03-01. March 5, 2026. FDA. fda.gov.

FDA, Warning Letter to Purolea Cosmetics Lab, FEI 3011669383 / Ref. 320-26-58. April 2, 2026. fda.gov

CIOMS Working Group XIV, Artificial Intelligence in Pharmacovigilance. Council for International Organizations of Medical Sciences. December 4, 2025. cioms.ch

EMA/HMA. Network Data Steering Group Workplan 2026–2028. Adopted February 2026. ema.europa.eu

Joint FDA/EMA document: Guiding Principles of Good AI Practice in Drug Development. FDA. Accessed August 19, 2026 

https://www.fda.gov/about-fda/artificial-intelligence-drug-development/guiding-principles-good-ai-practice-drug-development

National Commission into the Regulation of AI in Healthcare: research, engagement and call for evidence findings. 2026. Medicines and Healthcare products Regulatory Agency (MHRA). Accessed August 19, 2026. 

https://www.gov.uk/government/groups/national-commission-into-the-regulation-of-ai-in-healthcare

Guiding principles for the EMA–US FDA cluster teleconferences on artificial intelligence in pharmacovigilance. EMA/422032/2023. European Medicines Agency and US FDA. Published March 2, 2026. Accessed August 19, 2026. 

https://www.ema.europa.eu/en/documents/other/guiding-principles-ema-us-fda-cluster-teleconferences-artificial-intelligence-pharmacovigilance_en.pdf

Beena Wood is chief product officer at Qinecsa Solutions, where she leads product strategy across the company’s pharmacovigilance technology portfolio.

Articles

Regulatory and industry scrutiny of AI in pharmacovigilance has shifted from post hoc explainability toward prevention, enforcement, and longitudinal governance in post-marketing change management. Recent FDA warning letters highlight two ends of the same failure mode: unmonitored rule-based processes and undocumented delegation of compliance judgment to AI agents. Regulators are converging on lifecycle monitoring, risk-calibrated human oversight, and plain-language transparency to anchor accountability. Explainable outputs can paradoxically reduce reviewer skepticism, necessitating controls on human behavior and operational metrics. Readiness is defined by detectability of degradation, not narrative quality.

Post-marketing governance must be evidenced across the full lifecycle, design, validation, and live monitoring, not documented once and assumed to hold.

How AI Can Assist with Post-Marketing Changes

AI accelerates biopharma from discovery to production, cutting cycle times by up to 40%.

How AI is Transforming the Biopharmaceutical Value Chain from Discovery to Manufacturing

The biopharmaceutical industry is undergoing a systemic transformation as AI integrates into every stage of the value chain, from initial molecule discovery to large-scale manufacturing.¹ Current data indicate a significant compression of timelines, with researchers now designing new molecules in weeks rather than months. In the critical target identification phase, AI-driven workflows are yielding an average time savings of 28%.

Strategic investments underscore this shift, notably a $1 billion collaboration between NVIDIA and Eli Lilly.² This initiative aims to develop an AI factory for medicine, leveraging large-scale models trained on the language of biology and chemistry. Central to this approach is a continuous learning loop where physical wet-lab experiments and computational dry-lab models inform one another in real time.

Beyond discovery, the emergence of Pharma 4.0 is redefining production through smart factories. By utilizing real-time analytics, robotics, and sensor networks, facilities are transitioning from traditional batch production to continuous manufacturing.³ This shift has reportedly reduced production cycle times by 30% to 40% and enabled predictive maintenance to prevent batch failures. Efficiency gains also extend to infrastructure; one case study noted a 21% reduction in electricity usage for cooling systems through AI optimization alone.

Despite these technical advancements, the transition presents significant pedagogical and regulatory challenges.¹ Professionals must be trained to critically evaluate AI outputs to determine if a system is operating within its competency or hallucinating confidently. To manage these risks, regulatory bodies like the FDA and EMA have established 10 guiding principles focused on human-centric design and rigorous data governance. Ultimately, the technology serves as an augmentative tool rather than a total replacement for human expertise.

Editor's note: This transcript is a lightly edited rendering of the original audio/video content. It may contain errors, informal language, or omissions as spoken in the original recording.

From the absolute earliest stages of discovery all the way to the factory floor, artificial intelligence is completely changing how we create medicine. So how exactly is this all unfolding? Let's dive in and take a look. Artificial intelligence is fundamentally changing the timeline for drug discovery, which I think we're all seeing a lot of.We're watching researchers design new molecules in weeks rather than months, which is amazing to see, and it's really helping identify promising candidates with a precision that wasn't possible even two years ago. That is a total game changer for discovery timelines, and we're seeing real, tangible results. In that critical first step of finding a drug target, AI is delivering an average time saving of 28%. That's more than a quarter of the time just gone. And make no mistake, this isn't just happening in some small-scale experiments. We're seeing massive strategic investments. I mean, take this collaboration between Nvidia and Lilly. That is a $1 billion bet on building an AI-powered engine to completely reinvent drug discovery, and their goal is incredibly ambitious. They're essentially building an AI factory for medicine. They're creating these huge models that understand the very language of biology and chemistry, and the really cool part is this continuous learning system where the wet labs experiments, you know, the actual physical testing, and the dry lab computer models are constantly talking to each other, getting smarter together. So let's follow the journey from the lab to the production line. This is where the AI-driven smart factory, or Pharma 4.0, really enters the picture. So what is a smart factory? Well, think of it as a production facility that's just completely connected. It uses a whole network of sensors, robotics, and AI to make manufacturing way more efficient with higher quality and just a lot more agile than we've ever seen before. And again, you can see the impact in the real world. In one case study, Merck KGaA used an AI platform to optimize its cooling systems and cut their electricity use by 21%. And get this, they did it without any major changes to the physical equipment itself. I mean, these efficiency gains are just huge. By shifting to what's called continuous manufacturing, a process that's only really possible because AI can monitor everything in real time, some facilities are slashing their production cycle times by a massive thirty to forty percent. Predictive maintenance tools are catching equipment issues before they cause batch failures. Real-time analytics are enabling faster quality decisions, and even continuous processing is becoming more viable as digital monitoring removes some of that uncertainty that made manufacturers cautious about moving away from batch production. And this table really sums up the fundamental shift, doesn't it? We're moving away from the old way of doing things, which was often reactive and based on batches, to a new model that's continuous, it's predictive, and it's all driven by real-time data. It's about being proactive instead of reactive. Okay, but let's be real. This isn't just about plugging in some new technology and calling it a day. The real skills challenge over there is how are we teaching people to critically evaluate AI outputs in the context of the work? How are we teaching people to evaluate if AI is operating within competency or is it hallucinating confidently? How do we teach that? How do we teach when AI cannot answer a question? And the good news is the big regulatory bodies like the EMA and the FDA, they get this. They've actually come together to establish these ten guiding principles, creating a kind of playbook for using AI safely and ethically across the entire life cycle of a drug. And if you look at these principles, a really clear theme emerges, right? It's all about human-centric design, taking a risk-based approach, and having rock-solid data governance. The focus is absolutely on making sure these powerful tools are used responsibly with patient safety as the number one priority. And that really brings us to the final, and I think most powerful thought. The general consensus seems to be that, no, AI is not going to replace the expert. But the real question for all of us in this industry is this: will the experts who use AI replace the ones who don't?

European Medicines Agency; U.S. Food and Drug Administration. EMA and FDA set common principles for AI in medicine development. EMA. 2026. Accessed February 20, 2026. 

https://www.ema.europa.eu/en/news/ema-fda-set-common-principles-ai-medicine-development-0

NVIDIA and Eli Lilly and Company. NVIDIA and Lilly announce co-innovation AI lab to reinvent drug discovery in the age of AI. NVIDIA Newsroom. Published January 13, 2025. Accessed February 23, 2026. 

https://nvidianews.nvidia.com/news/nvidia-and-lilly-announce-co-innovation-lab-to-reinvent-drug-discovery-in-the-age-of-ai

Merck KGaA. Merck cuts cooling energy by 21% with AI optimization from etalytics. etalytics. Published February 11, 2026. Accessed February 23, 2026. 

https://etalytics.com/resources/blog/merck-reduces-cooling-energy-use-with-ai-optimization-from-etalytics