The Early Development Safety Roadmap (EDSR) is a critical framework in pharmaceutical research, designed to proactively identify, assess, and manage safety risks from preclinical stages through Phase 1 and 2 clinical trials. Authoring EDSRs is challenged by fragmented data sources, evolving templates, and complex cross-functional collaboration. This presentation explores how Artificial Intelligence (AI), specifically ChatGPT, can transform EDSR creation by streamlining repetitive tasks, integrating diverse datasets, and automating initial draft generation. We highlight best practices in data preparation, prompt engineering, and quality review processes, emphasizing the collaborative role of pharmacovigilance (PV) physicians and scientists. Demonstrations showcase how ChatGPT interfaces with document management systems and literature databases, accelerating literature searches, information summarization, and reference list generation. Future directions include deeper automation via ChatGPT API and copilot solutions, enabling efficient template-based drafting and seamless PV team engagement. By leveraging AI-powered tools, organizations can improve the quality, consistency, and speed of safety assessments, supporting robust risk mitigation strategies in early drug development.

Automation is already reshaping pharmacovigilance — quietly, compliantly, and effectively. This session explores why automation, not AI, is the real innovation in safety-critical workflows. We’ll discuss regulatory expectations, current use cases, and how rule-based systems deliver reliable, auditable results. Whilst AI remains aspirational, automation is operational — and inspection-ready!

Objectives include:

• Understand the key differences between automation and AI in pharmacovigilance.
• Learn where automation is already successfully used in PV processes.
• Explore regulatory expectations around validated systems and inspection readiness.
• Assess risks of unvalidated AI versus the compliance benefits of automation.
• Discover how hybrid models can balance automation with human oversight.

Sponsored by PharSafer.

This presentation describes a novel AI-driven system designed to support the authoring of Periodic Benefit-Risk Evaluation Reports (PBRERs), a critical component of pharmacovigilance. The system leverages generative AI, natural language processing, and machine learning to analyze safety data, generate draft content, and ensure alignment with regulatory expectations. Early prototypes have shown promising results in reducing manual workload and improving quality, consistency, and traceability in reporting. The presentation will share key technical insights, challenges encountered, and the potential impact on drug safety operations.

In this presentation, we will be summarized in 20min our 10 years of experience and lesson learnt transforming the PV operations with next generation platform.

During the session we will discuss the challenges, opportunities, new technological advances, of the AI-driven platforms. The audience will have the unique opportunity to interact with first hand experienced TCS team and learn about how the current PV landscape can be totally reimagine. Whether you’re excited by AI’s possibilities or curious about how it integrates with current practices, today’s discussion will offer valuable insights into this evolving field.

Pharmacovigilance (PV) case intake is critical in adverse event (AE) reporting, ensuring compliance with global regulatory requirements and maintaining patient safety. However, for many years, the increasing complexity and volume of individual case safety reports (ICSRs) have presented challenges for organizations using manual workflows. Traditional case intake methods, which include receiving, transcribing and validating reports, were becoming increasingly unsustainable as regulatory expectations evolved and the volume of cases increased. Despite the optimism shown by some technology vendors that the use of artificial intelligence (AI), there was scant evidence of significant deployment of AI in real functioning pharmacovigilance systems over the years. Barriers to adaptation were attributed to significant costs of the system configuration and datasets needs for both implementation and ongoing maintenance.

In this presentation, PPD™ Functional Service Partnership (FSP) Pharmacovigilance solutions experts:

• Discuss barriers to adoption of earlier systems and two key factors which have significantly reduced these barriers over the past four years.
• Re-examine the benefits of AI-assisted case intake and processing.
• Share five key implementation considerations.

Sponsored by PPD clinical research business of Thermo Fisher Scientific.

Sponsored by Techsol Life Sciences.

Discover how a generative AI tool was designed to streamline pharmacovigilance workflows by extracting and organizing adverse event data. This session shares key lessons from development to deployment—including human-in-the-loop strategies and transparent communication—that helped earn user and stakeholder trust. Gain a practical framework for responsibly implementing GenAI in drug safety.

Sponsored by Syneos Health.

Sponsored by Indegene.

Attendees will gain a deep understanding of how AI, machine learning, and automation are revolutionizing pharmacovigilance by streamlining case processing, enhancing signal detection, and driving operational efficiency. They will explore real-world applications, innovative tools, and forward-looking strategies to build trust in AI and future-proof safety operations across the drug development lifecycle.

Discover how a small pharma company is reshaping pharmacovigilance with AI.In a world where 95% of AI pilots fail to deliver ROI, Wörwag Pharma and AAA-Pharma have defied the odds. This session unveils their journey from skepticism to success—leveraging theralyze.ai to automate E2B R3 case generation, reduce manual workload, and maintain regulatory compliance through human quality control. If you're curious about real-world AI implementation in safety databases, this is a story you won’t want to miss.

From proof-of-concept to practical impact—fast.Learn how rapid development, agile feedback loops, and a commitment to quality enabled a responsive AI system to evolve week by week. Reinhold Schilling shares firsthand insights into overcoming validation challenges, managing risks, and transforming case intake processes. Whether you're in pharma, tech, or compliance, this session offers valuable lessons on making AI work in high-stakes environments.

The pharmacovigilance (PV) landscape is evolving rapidly with advancements in artificial intelligence (AI), transitioning from traditional Generative AI (Gen AI) to the more autonomous Agentic AI. This presentation, titled "Gen AI to Agentic AI: Multiplying PV Writing," explores how this paradigm shift enhances the efficiency, accuracy, and scalability of PV documentation, including Periodic Benefit-Risk Evaluation Reports (PBRERs), Risk Management Plans (RMPs), and adverse event narratives. Gen AI has already revolutionized initial draft generation by leveraging natural language processing to produce high-quality content from structured data. However, Agentic AI takes this further by autonomously managing end-to-end workflows—identifying data gaps, cross-referencing regulatory requirements, and iteratively refining outputs—effectively multiplying productivity while minimizing human oversight. Through case studies, we demonstrate a 40% reduction in writing time and a 25% improvement in compliance adherence when transitioning from Gen AI to Agentic AI systems. Attendees will gain insights into implementation strategies, challenges (e.g., data quality, validation), and future potential, equipping PV professionals to harness AI-driven automation for enhanced safety reporting in a regulatory landscape.

Sponsored by Quartica.

This fireside chat will deliver a compelling narrative around the potential of AI based tools, focusing on the real (human) foundations of deployment like cultural change and building trust.

Sponsored by ClinChoice.

Artificial intelligence (AI) holds significant promise for strengthening proactive drug safety measures throughout the product lifecycle, spanning pre-clinical research, clinical trials, and post-market surveillance. Regulatory bodies worldwide recognize AI as a powerful tool capable of identifying, monitoring, and mitigating safety risks, such as FDA’s Emerging Drug Safety Technology Program (EDSTP). Drug developers are also poised to harness the power of AI to achieve the best safety profiles for products and reduce development costs.In pre-clinical research, AI-powered quantitative structure-activity relationship (QSAR) models can predict toxicity, helping prevent unsafe compounds from entering clinical trials. In clinical development, AI can help further identify drug characteristics leading to potential adverse events, predict possible drug–drug interactions, and identify high-risk populations. In post-market phases, AI can be applied in adverse drug event detection and automated processing of safety reports. Context-aware machine learning solutions can support pharmacovigilance (PV) departments by automating repetitive tasks, analyzing vast volumes of real-world data, and providing decision support for benefit-risk discussions and causality assessment of safety events. However, several challenges must be addressed. Limitations in data quality and consistency impedes robust development of AI technology. AI technology still needs to mature in areas such as transparency, interpretation, and validation. As a result, uncertainties remain for integrating AI-driven outputs into official safety assessments.Moving forward, collaboration among industry, academia, and regulatory agencies is vital to fully realize AI’s potential in strengthening drug safety. With effective governance, robust datasets, and ongoing methodological innovation, AI can help usher in a new era of efficient, data-driven pharmacovigilance and safer therapeutic developments worldwide.

Abstract: First impressions are everything, especially during a Pharmacovigilance Inspection. As the Inspector’s entry point into an organisation’s PV system, the Pharmacovigilance System Master File (PSMF) can be leveraged to showcase continuous improvement in key areas and demonstrate a commitment to patient safety through regulatory compliance.

With real-world examples and practical tips, attendees will learn the role the PSMF plays during Pharmacovigilance Inspections, current best practices and how to anticipate inspector queries.

By adopting a proactive approach to PSMF management, organisations can transform this standard regulatory requirement into a strategic asset that drives transparency, efficiency, and compliance to give the perfect first impression.

Sponsored by ProPharma.

Attendees will gain a deep understanding of how AI, machine learning, and automation are revolutionizing pharmacovigilance by streamlining case processing, enhancing signal detection, and driving operational efficiency. They will explore real-world applications, innovative tools, and forward-looking strategies to build trust in AI and future-proof safety operations across the drug development lifecycle.

To be announced.