Optimizing Financial Operations with LLMs: CFO’s Perspective
summary
Claims a deployed LLM pipeline cuts fraud and false positives — significant drop in fraudulent transactions — by analyzing transactions plus message text; integrates ERP data, supplier sites and news to produce near realtime cashflow updates that let teams reduce idle cash and redeploy capital; requires months of data cleaning and uses explainability tools that surface contributing data for stakeholder trust; prototypes autonomous finance actions like autonomous invoice approvals under rule scaffolding and human fallback, with pilots showing promising accuracy.
AI-written summary · about
chapters (10)
- 0:00Welcome to Con 42 Large Language Models 2025
- 1:30The Financial Landscape Before LLMs
- 3:37The Power of LLMs in Finance
- 5:37Fraud Detection with LLMs
- 8:22Transforming Cashflow Management
- 10:32LLMs in Legal Compliance
- 12:36Challenges and Hurdles
- 16:13The Future of Finance with LLMs
- 17:56The Evolving Role of the CFO
- 19:15Conclusion and Final Thoughts
related talks
36mLLM-Driven Financial Ledgers: Real-Time Accounting Systems at Scale
44mPanel: LLM use in business transformation
31mWhat problems should enterprises actually trust an LLM to solve?
39mLeveraging AI and Emerging Technologies for Competitive Advantage
21mHyperautomating Compliance and Risk Intelligence in Finance
29mAutomating Privacy Reviews with LLMs: Challenges & Opportunities
33mDesigning LLM Applications for Massive Datasets
26mAI Agents as Financial Co-Pilots | AI-Driven Strategy & Decision Making