AWS FSI Executive Roundtable: Trust in the Age of AI

​From Probabilistic to Predictable: Taming LLM Hallucinations for Regulated Workloads

​In the rapidly evolving landscape of financial services, artificial intelligence stands as both a transformative opportunity and a formidable challenge. Our AWS FSI Executive Roundtable: Trust in the Age of AI brings together industry leaders, innovators, and practitioners to navigate this complex terrain.

​This isn't your typical high-level AI discussion. We're diving deep into the trenches of real-world implementation, addressing the unique hurdles faced by financial institutions as they harness the power of Large Language Models and other AI technologies. From regulatory compliance to client trust, we'll tackle the hard questions that keep FSI executives up at night.

​Expect frank discussions on model reliability, the intricacies of fine-tuning versus other approaches, and the delicate balance between innovation and risk. Our carefully curated group of speakers will share battle-tested strategies, unexpected pitfalls, and breakthrough solutions.

​Whether you're on the front lines of AI implementation or charting your organization's AI strategy, this roundtable offers invaluable insights. You'll leave with concrete ideas, practical frameworks, and a network of peers facing similar challenges.

​Join us as we build the roadmap for trustworthy AI in financial services. The future of finance is being written now – be part of the conversation that shapes it.

​This round table will be followed by a happy hour.

Full Agenda

​Arrival & Peer Networking

​15:00 - 15:30 EDT

​Welcome & Keynote: The Challenge of Hallucinations in Financial Services AI

​15:30 - 15:45 EDT

​Introduction to the critical challenges facing financial institutions in implementing AI solutions, focusing on the need for deterministic outputs in regulated environments. Discussion of how hallucinations pose significant risks in financial decision-making and regulatory compliance.

​The Practitioner's Playbook: Non-Generative Fine-Tuning for Financial Services

​15:45 - 16:15 EDT

​Technical deep-dive into practical implementation strategies:

​1. Non-Generative Output Framework (Dr. Andy Li from Bank of America):

• ​    Yes/No answer implementation

• ​    Comparative analysis between companies

• ​    Bridge questions with direct quotations

• ​    Synthetic Q&A data generation

​2. Safe and Reliable use of Generative AI on AWS with Artificial Genius (Dr. Paul Bruchard from Artificial Genius):

• ​    Model configuration and setup

• ​    Training parameters optimization

• ​    Integration with SageMaker

• ​    Testing and validation approaches

​3. Evaluation Framework (Dr. Igor Halperin from Fidelity):

• ​    Measuring hallucination reduction

• ​    Testing multi-paragraph comprehension

• ​    Validating factual accuracy

• ​    Performance metrics for regulated environments

​Moderated Roundtable Discussion

​16:15 - 17:15 EDT

​Ilan Gleiser

​Focused discussion on:

• ​Real-world implementation challenges in financial services

• ​Regulatory compliance considerations

• ​Integration with existing workflows

• ​Best practices for model risk validation

• ​Scaling solutions across organizations

​Closing Remarks & Key Takeaways

​17:15 - 17:30 EDT

​Summary of key implementation strategies, lessons learned, and future directions for trustworthy AI in financial services.

​Happy Hour & Continued Conversation

​17:30 - 18:30 EDT

Speakers

​Ilan Gleiser Principal World Wide GenAI Specialist, AWS

​Principal GenAI Specialist at AWS. Expert in scalable AGI architectures and foundation model customization.

​Dr. Igor Halper is Group Data Science Leader at Asset Management Technology, Fidelity Investments

​Group Data Science Leader at Fidelity, ex-JPMorgan & ex-Bloomberg, author of ML in Finance, PhD physicist, RISK Buy-Side Quant awardee.

​Dr. Andy Li (PhD, CFA, FRM) GenAI Leader & Director at Bank of America

​16+ yrs in Data Science & AI. Leads ML/GenAI risk at BofA. ex-Amazon. Fordham lecturer. Expert in AI risk, model governance & responsible AI adoption.

​Dr. Paul Bruchard Cofounder and CTO at Artificial Genius, Inc.

​Cofounder and CTO at Artificial Genius, Inc. Ex-Goldman Sachs. Multidisciplinary innovator with 20 yrs leading teams to deliver transformative ideas.