This whitepaper explores recent advances in generative artificial intelligence (AI) and its potential to automate the extraction, interpretation and digitization of legal clauses from ISDA’s credit support annexes (CSAs) into standardized, machine-readable code using the Common Domain Model (CDM). Specifically, the paper benchmarks and reviews the ability of generative AI to accurately extract five CSA clauses and digitize them into CDM format. The findings suggest that integrating industry-specific data significantly boosts generative AI accuracy, larger generative AI models typically handle nuanced legal language better, and generative AI can be utilized within a modular framework (eg,agentic AI) to extract legal clauses more accurately and efficiently.
Click on the attached PDF to read the full report.
Documents (1) for Benchmarking Generative AI for CSA Clause Extraction and CDM Representation
Latest
ISDA Guidance: SOFR Publication on Good Friday 2026
ISDA guidance for parties to over-the-counter derivative transactions affected by expected non-publication of SOFR on Good Friday in 2026. Please note that the guidance may be updated from time to time.
ISDA Paper on FRTB Rules in Brazil
On March 24, ISDA submitted a paper to Banco Central do Brazil’s (BCB) on its implementation of the revised market risk framework under the Fundamental Review of the Trading Book (FRTB), which represents an important step toward strengthening prudential standards...
IQ Interview with Mark Uyeda
Mandatory clearing of US Treasury securities is due to begin at the end of this year under rules finalized by the Securities and Exchange Commission (SEC) in 2023. SEC commissioner Mark Uyeda talks to IQ about the benefits of clearing...
Response to FCA on CFI Codes for Transparency
On March 19, ISDA responded to Chapter 3 of the UK Financial Conduct Authority’s (FCA) Quarterly Consultation CP26/8 on transparency requirements for financial instruments under Market Conduct Sourcebook (MAR) 11. Sections 3.11-3.13 of the consultation paper explain a discrepancy between...