We convened a group of Medical Affairs experts from across the Pharma industry to explore the true reality of utilising AI for medical conferences. Our goal was to cut through the hype, assess where we are on the AI adoption journey, and share honest perspectives on what’s working, what’s not, and what lies ahead.

The roundtable opened with a candid assessment of where we are now. Words to describe AI included “revolutionary”, “game-changing” but also “scary”. One attendee noted that “AI is not the future. It is the present, and it’s fast”. But while AI has dominated headlines and high-level vision decks, the reality on the ground is more nuanced.

“AI is not the future, it is the present, and it’s fast”

This explains why the moment of broad excitement seems to be giving way to a more strategic approach, with teams stepping back to reassess and question: Where is AI truly adding value? Which pilots deserve deeper investment? And how can we build sustainable, scalable AI strategies rather than one-off experiments?  

Real-life use cases of AI across the congress lifecycle 

What was clear from our discussions is that medical teams are already experimenting with using AI to support with congress-related activities. The group showed varying levels of maturity, with some just beginning to explore AI’s potential, while others are actively piloting congress-specific tools. Perhaps unsurprisingly, none claimed to be fully integrating AI across the congress workflow, which reflects what we’re seeing across the broader pharma industry.

During the roundtable, participants shared how they are actively exploring and applying AI tools across the congress lifecycle:

1. Planning: Smarter preparation through AI

Most of the discussion around planning focused on how AI is being used to support agenda development, session prioritisation, and stakeholder identification. Teams are experimenting with tools like ChatGPT, Microsoft Copilot, and AI agents to streamline congress preparation. These tools help to sift through dense congress agendas, highlight high-priority sessions, and identify key opinion leaders or relevant stakeholders and are reducing manual effort and accelerating decision-making ahead of the event

2. Execution: Enhancing efficiency and onsite experience

Several teams are using AI tools like Copilot and AI-assisted notetaking platforms to capture and summarise discussions in real time. Voice-to-text capabilities are proving especially useful for increasing efficiency, potentially reducing the need for large onsite teams by enabling remote access, transcription, and analysis of session content. Some organisations are also beginning to explore AI-powered tools at medical booths, such as medical information chatbots that can handle common queries from congress attendees 

However, the group raised an important caution: there is a fine line between smart data capture and surveillance. There were concerns that attendees may perceive overly intrusive tools as “Big Brother”-like. Thoughtful implementation will be essential to ensure acceptance and avoid undermining engagement. 

3. Reporting: Accelerating insights and personalising follow-up

In the post-congress phase, participants highlighted how AI is already helping to generate rapid insights and synthesise large volumes of data into concise, actionable summaries. This is speeding up internal reporting and enabling faster follow-up. The group also discussed the use of AI to coach MSLs on tailoring post-congress engagement with HCPs to allow for more personalised follow-up. Finally, the group highlighted the potential for AI to measure congress ROI, although this requires further exploration.  

The implementation gap: Barriers to AI adoption 

Many Medical Affairs teams reported experiencing ‘analysis paralysis’, feeling overwhelmed by the sheer number of potential AI applications, uncertain where to invest, or concerned about making the wrong choice. Others moved quickly into pilots, but without a clear understanding of what success should look like from the outset. In many cases, teams lacked the frameworks, benchmarks, or internal alignment needed to define meaningful outcomes, making it difficult to evaluate what worked and why, ultimately leading to doubts around ROI of AI.  

During our discussion, three core themes emerged as barriers to meaningful AI adoption: 

• Capability gaps: Including varying levels of AI maturity across teams, limited training resources, and knowledge gaps in understanding AI functionalities, interpreting outputs, and evaluating tool effectiveness. These challenges are compounded by the difficulty of tracking tool usage and effectiveness consistently across global and local levels, making it harder to share best practices or scale success 

• Opportunity constraints: Lack of time and budget, integration issues with existing systems, siloed or restricted access to high-quality data, insufficient buy-in from leadership, compliance restrictions and data security concerns continue to hinder implementation  

• Motivational blockers: Including skepticism around AI’s tangible benefits, fear of making the wrong investment, and resistance to disrupting familiar ways of working. In some cases, there’s apprehension about job displacement or losing control over decision-making