Document Processing
Merck needed to remove complexity and manual effort from the clinical trial process, where medical writers spent months reviewing thousands of pages of clinical data to produce regulatory-quality clinical study reports.
Reported impact
Merck shortened the creation of fully human-reviewed first drafts of clinical study reports, improved draft quality, and accelerated the path to making medicines available to health care providers and patients.
Merck built an internal generative AI platform to accelerate authoring of clinical study reports (CSRs) within its clinical development process. The implementation combines advanced table pre-processing with large language models (LLMs) and a proprietary gen-AI authoring platform. Clinical study data and tables are mapped, extracted, styled, and validated before the LLM-driven authoring step generates a high-quality first draft for qualified medical writers to review. Merck also changed its operating model around the platform, training teams in prompt engineering and data pipelines so they could supervise outputs and manage the workflow safely. The system is designed for rigorous human oversight rather than autonomous submission, and it has already been used across multiple studies and in live CSR submissions. Merck is now scaling the platform across its late-phase pipeline.
Clinical study data and tables are ingested into Merck's proprietary platform, where advanced table pre-processing handles mapping, extraction, styling, and validation. That prepared content is passed to LLM-based gen AI authoring to generate a CSR first draft, which is then reviewed by qualified medical writers before use in Merck's clinical study reporting workflow.
"In as little as five minutes, it can deliver a high-quality first draft, managing table mapping, data extraction, styling and validation."
"Increased the quality of CSR drafts – as measured by reducing the number of errors by 50%"
"Merck has implemented a new internal generative-AI-powered platform that significantly reduces the time needed to produce critical-path clinical study documents"
"Merck technologists and clinical trial experts have developed new solutions leveraging large language models (LLMs) to expedite the creation of first drafts of clinical study reports (CSRs)"
"Reduced the time to create a fully human reviewed CSR first draft from an average of 180 hours to 80 hours."
"the co-developed platform integrates advanced table pre-processing with gen AI authoring"
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
In-house Sample Jul 28, 2026 Novartis Novartis R&D needs to turn large volumes of scientific, clinical, operational, and real-world data into faster, higher-quality decisions across drug discovery and development, reducing cycle times, surfacing safety… less than 10 percent of drug candidates historically reach the market AI models · AI-enhanced analytics · Generative AI In-house Sample Apr 13, 2026 Elevance Health Elevance Health needed to find fraud, waste, and abuse in very large volumes of claims, payment, contract, and medical-record data so it could detect billing errors, suspicious coding patterns, and improper payments… across an annualized volume of nearly 23,000 reviews In-house Sample Jun 19, 2025 Sanofi Sanofi needed to replace siloed, manual manufacturing and supply workflows with a connected, data-driven operating model that could improve medicine quality and safety, prevent supply disruptions, speed product… 3,500 annual Product Quality Reports Modulus · advanced analytics · generative AI 