Large language models (LLMs) are creating new opportunities across pharmaceutical research, but their outputs are only as trustworthy as the data and context supporting them. AI-generated responses can sound scientifically plausible while still being inaccurate, making reliable grounding particularly important in R&D.
In this session, Alice Augustine of GSK and Jim Morris of Progress Software examine how semantic grounding can help address this challenge. The webinar explores GSK's approach to classifying data using established ontology classes and replacing ambiguous free-text references to key research entities with persistent Uniform Resource Identifiers (URIs).
Watch the on-demand webinar to explore:
Why plausible-sounding AI outputs can still create reliability challenges in pharmaceutical R&D
How semantic grounding provides structured context for validating AI-generated outputs
How GSK uses ontology classes to transform unstructured “strings” into well-defined “things”
Why persistent URIs can help replace ambiguous free-text references to research entities
How a structured ground truth can support data integrity and more trustworthy AI systems
Jim Morris
Progress Software
Alice Augustine
GSK
Log in and download any resource without filling out the forms. Log in with ProgressID
OR
Related Resources
