Continental Resources, PwC leverage generative AI to optimize oil & gas ops
summary
Continental Resources and PwC describe a production generative AI rollout that starts with a health and safety assistant pulling PDFs, standards and bulletins into a knowledge store and exposing clickthrough transparency on mobile, plus a text-to-SQL well query assistant that lets non‑analysts ask plain English. Implementation uses Bedrock models (Claude and Amazon Titan for embeddings), DataBricks/Spark pipelines, KMS and CloudWatch, plug‑in Lego architecture, 6–8 week sprints, load tests at ~80–100 concurrent users and built‑for‑reuse components with business feedback loops.
AI-written summary · about
chapters (14)
- 0:00Introduction and speakers
- 0:44Why generative AI now
- 3:09Generative vs predictive AI
- 6:00First use case health and safety
- 8:28Why PWC and AWS partnership
- 11:21Project goals and use cases
- 13:52Plug-in architecture and Lego blocks
- 16:22Next roadmap and innovation
- 19:19Addressing knowledge gaps
- 21:03Measuring success and milestones
- 23:52AWS services and benefits
- 27:03PWC’s value for partners
- 32:42Future of generative AI in energy
- 34:50Closing remarks
related talks
30mGenerative AI in action: from POC to business value
27mRio Tinto unlocks potential of generative AI
28mImproving responses in generative AI applications
18mNavigating the growth frontiers of generative AI
20mExplore what’s possible with new AWS generative AI services
23mGenerative AI use cases
58mCognizant cognitive architecture for generative AI: The path to MVP
18mAccelerate generative AI development with integrated data platforms