Project Management with NotebookLM for Clinical Teams: Where AI tools meet Execution.

A Let’s Talk session on 30th June 2026

Moderator: Dr. Volker Moeckel

Slide set

Project Management with NotebookLM_2.pdf

Download PDF • 904KB

Meeting Notes:

1. Introduction to Notebook LM for Clinical Project Management: Volker introduced Notebook LM (LM=Language Model) as an AI-powered research assistant for clinical project management, highlighting its secure, source-grounded approach and its potential to transform document analysis and workflow in clinical trials, with participants engaging in the discussion.

   1. Notebook LM Overview: Volker explained that Notebook LM is an AI tool from Google designed to serve as a secure, source-grounded project knowledge base, capable of analysing up to 50 documents such as protocols, regulatory guidelines, and meeting notes, and providing citation-backed answers without referencing external data.

   2. Document Input Flexibility: Notebook LM accepts a variety of input formats, including Office files, URLs from videos, podcasts, and HTML pages, allowing clinical teams to consolidate diverse sources for analysis and extraction of relevant information.

   3. Role-Based Output Customization: Users can define their role within Notebook LM, such as project manager or scientific advisor, enabling the tool to tailor its responses and outputs to the specific needs and context of the user.

   4. Output Formats and Features: Notebook LM can generate multiple output types, including text summaries, PowerPoint presentations, PDFs, podcasts, and infographics, providing flexibility for clinical teams to communicate and share information in various formats.

2. Use Cases and Practical Applications of AI Tools in Clinical Trials: Volker led a discussion on practical use cases for Notebook LM and similar AI tools in clinical trial management, with participants sharing their experiences and perspectives on protocol analysis, regulatory compliance, meeting note tracking, and collaborative workflows.

   1. Protocol Analysis: Notebook LM can manage the upload of complex study protocols, enabling instant queries for inclusion/exclusion criteria, drug dosage schedules, and safety reporting procedures, even when multiple amendments are present.

   2. Regulatory and Safety Surveillance: The tool can generate checklists from regulatory guidelines such as ICH and 21 CFR Part 11, identify required safety updates, and compare trial documents against regulatory requirements to ensure compliance.

   3. Cross-Study Synthesis: Notebook LM can consolidate data from multiple clinical studies and diverse sources, including YouTube videos and other documentation, to identify discrepancies, bottlenecks, and facilitate comprehensive analysis.

   4. Meeting Note Tracking: By uploading meeting transcripts, clinical teams can use Notebook LM to track project progress, verify if action items have been addressed, and analyze management satisfaction with ongoing tasks.

   5. Audit Preparation and Collaborative Workflows: Notebook LM can assist in preparing for internal audits by generating FAQs and briefing documents, and supports collaborative workflows by allowing sharing of notebooks within teams, enabling collective review and question-answering about trial data.

3. Challenges and Limitations of AI Tools in Clinical Project Management: Participants discussed the challenges and limitations of AI tools such as Notebook LM, focusing on document accuracy, data protection, document limits, and the irreplaceable role of human interaction in project management.

   1. Accuracy and Trustworthiness: Concerns were raised about the accuracy of AI-generated outputs, noting that even with a single document, AI tools like ChatGPT, Claude, or Notebook LM may not fully understand the content, prompting questions about reliability when uploading large numbers of documents.

   2. Data Protection and Security: The importance was emphasized of protecting proprietary information, describing their companies' use of firewalls and secure systems to prevent sensitive data from leaving the organization, and highlighting the need for closed systems like Notebook LM.

   3. Document Limitations: It was mentioned that the presumed limitation of Notebook LM's 50-document cap per notebook might pose a challenge for clinical trials that often require managing more documents, and discussed the potential for future versions or enterprise solutions to address this issue.

   4. Human Interaction and Communication: It was agreed that AI tools cannot replace human interactions, communication, and understanding of individual circumstances, which remain essential for effective project management and team collaboration.

4. Best Practices and Skills for Effective AI Tool Usage: Best practices for using AI tools were discussed, including prompt engineering, role definition, and the importance of developing both AI and human 'power skills' for future project managers.

   1. Prompt Engineering: An example was shared where a consultant used a detailed, multi-page prompt to generate a three-year strategy with Claude (AI tool), demonstrating that the quality of AI outputs depends heavily on the specificity and depth of prompts.

   2. Role Definition: It was highlighted that defining the user's role within the AI tool, such as acting as an expert advisor, can improve the relevance and quality of responses.

   3. Power Skills vs. AI Skills: The need for project managers to develop both AI skills and 'power skills' was emphasized, such as relationship building, conflict resolution, and team management, which AI cannot replicate.

5. Future Directions: Agentic AI and Portfolio Management: The potential of agentic AI tools for portfolio management was explored, discussing how standardized data and collaborative workflows could enable AI agents to autonomously analyze project performance and support decision-making.

   1. Agentic AI Potential: The next wave of AI tools was described as agentic, capable of acting autonomously on outputs if authorized, and suggested that future sessions may focus on their impact in project management.

   2. Portfolio Management Applications: Participants envisioned agentic AI assisting in portfolio management by analyzing standardized project data, identifying risks, and supporting resource allocation and decision-making, provided data structure and capture challenges are addressed.