How to deliver business value in financial services with GenAI
summary
Ergo builds an AI Factory on AWS, ingests PDFs and website content into an Ergo GPT front end backed by a vector database (AWS OpenSearch), and uses OCR via AWS Textract. Fifty internal use cases get screened; 32,000 employees roll out Ergo GPT in waves. External A/B testing with 500 customers each shows positive feedback rising from 36% to 46%. Early hallucinations spike to 177% but drop to 4% after data cleaning and curation; Llama 2/3, Claude 3 and OpenAI are used.
AI-written summary · about
chapters (7)
- 0:00Introduction and welcome
- 3:33Speaker introductions and company overview
- 8:13House of Generative AI strategy
- 17:24Customer-facing GenAI chatbot deployment
- 23:48AI Factory and governance for GenAI
- 31:59MLOps for GenAI applications
- 37:22Closing remarks and Q&A
related talks
20mExplore what’s possible with new AWS generative AI services
27mBuilding Generative AI features
51mTransforming Banking with Advanced AI: GenAI, RAG, and Multi-Agent Systems
28mIs GenAI the technology revolution we need? How can we harness it at work?
31mGenAI after the Hype: Transforming Organizations with GenAI-based Agents
09mGen AI for regulated industries
02h 35mBringing GenAI to the Modern Enterprise. A production use-case. In Serverless Java!
24mInfuse Generative AI in your apps using