For the longest time, companies stayed away from trying out new tech stacks in pharmacovigilance (PV). But with the rise of generative AI, things are changing-fast.

Big players like Bayer teamed up with Genpact to use AI for automating case intake. Sanofi joined hands with IQVIA to apply AI for end-to-end case processing. And these are just a couple of examples. As case volumes keep going up year after year, everyone’s now in a race to bring AI into the game-to cut costs, boost quality, speed up case processing, and let safety teams focus on what really matters.

Applications of AI in Pharmacovigilance

Regulators Are Getting Involved Too

It’s not just pharma companies and tech vendors jumping on the AI bandwagon. Regulatory bodies are also gearing up for AI in PV and working with CROs and pharma companies to draft AI usage guidelines. Here’s what’s happening:

  1. FDA introduced CDER Emerging Drug Safety Technology Program (EDSTP) to explore the integration of artificial intelligence (AI) and other emerging technologies into pharmacovigilance (PV) activities. Through EDSTP, FDA wants to facilitate dialogue between CDER and industry stakeholders to discuss AI applications, enable knowledge transfer and build regulatory guidelines and policies. Read more here
  2. CIOMS Working Group XIV published draft guidelines for the use of artificial intelligence or intelligence augmentation in the field of pharmacovigilance. Public can review and provide feedback by 6th June 2025. The guidelines cover broad range of topics ranging from risk based approach, human oversight, validating AI, transparency, data privacy, fairness and governance model. Read more here
  3. EMA published a reflection paper last year, which talks of applications within PV for classification and severity scoring of adverse event reports. Read more here

Case Intake Automation

I was reading this great paper - Artificial intelligence in pharmacovigilance - Opportunities and challenges - and one quote really stood out:

“Full automation of PV system needs to consider two aspects – people and processes. The focus should be a collaborative approach of technical expertise (people) combined with intelligent technology (processes) to augment human talent that meets the objective of the PV system and benefit all stakeholders. AI technology should enhance human intelligence rather than substitute human experts.” That’s a big point the whole industry seems to agree on: AI should augment, not replace, case processors. Keep humans in the loop.

Most of the real AI use I’ve seen so far is around case intake automation. Tools can now automatically read source files—structured or unstructured—and create cases in minutes instead of hours. For example, Yeza.ai intake solution can read everything, even handwriting and multi-page lab reports, in just a couple of minutes. Pretty cool, right? Companies are also using AI to extract medication reviews from social media, encode using MedDRA and book cases. Another interesting use case was NLP query tool, where end users can talk to the safety DB in natural language. A year ago, who would’ve imagined that?

Beyond Just Case Intake

AI is evolving so fast, I’m pretty sure this post will feel outdated in no time. But here are a few things I can imagine AI handling in safety systems very soon:

  1. System interoperability without standard formats – Source systems might soon push data directly into safety systems without needing to generate E2B files.
  2. AI-generated narratives – AI could write better, more detailed narratives by reviewing lab data and literature sources.
  3. Drafting aggregate reports – Imagine AI going through thousands of cases to create summaries and draft reports—saving months of work.
  4. Intelligent MedDRA Coding – AI can analyze text and suggest the most relevant MedDRA terms. Fun fact: Bayer’s already doing this.
  5. PV + Regulatory Intelligence Integration – AI could help spot gaps in configuration, like incorrect reporting rules or missing business configurations.

😃

That’s all for now! Thanks for reading. Let me know what you think or what you’re seeing on your side of the PV + AI world.