BIS Bulletin Examines Cognitive Limits of Large Language Models
The use cases of generative AI in the banking sector are evolving fast, with many institutions adopting the technology to enhance customer service and operational efficiency. However, this adoption has been accompanied by increasing regulatory scrutiny, as regulators seek to address the complexities and risks associated with generative AI. These models’ potential for biased decision-making is one of the key concerns regulators have and thus is also increasingly an area of research. A recent Bulletin from the Bank of International Settlements (BIS) discusses the cognitive limits of large language models (LLMs) such as Generative Pre-trained Transformer (GPT). The authors, Fernando Perez-Cruz and Hyun Song Shin, explore whether these models truly understand the content they generate or merely replicate the text encountered during training.
The authors tested GPT-4 with a logic puzzle known as Cheryl’s birthday puzzle. The authors highlight the impressive capabilities of LLMs, including generating computer code, images, and solving complex mathematical problems. The LLM solved the puzzle flawlessly when presented with the original wording but failed when small incidental details, such as the names of the people, the number of dates, or the order of the clues, were changed. This suggests that while LLMs can perform tasks flawlessly when presented with familiar wording, they struggle with tasks that require rigorous reasoning and understanding when the wording is changed. The authors argue that this failure indicates a lack of true understanding of the underlying logic and a reliance on familiar wording. They also highlight the model's lack of self-awareness of its own ignorance.
The authors note that central banks’ activities are well-suited for the application of machine learning and artificial intelligence, reflecting the ample availability of structured and unstructured data, coupled with the need for sophisticated analyses to support policy. Historically, central banks had been early adopters of machine learning methods in statistics, macroeconomic analysis, and regulation/supervision. The authors also highlight that these findings do not detract from the tangible and rapid progress being made in these areas, as well as in scientific applications of artificial intelligence that have seen rapid progress. However, the authors conclude that while LLMs have made considerable progress in applications such as data management, macro analysis, and regulation/supervision, caution should be exercised when deploying them in contexts that demand rigorous reasoning in economic analysis. While the economic impact of AI could be significant, the current generation of LLMs falls short of achieving artificial general intelligence and cannot substitute for the rigorous reasoning abilities necessary for some core analytical activities. However, the ability of LLMs to engage in rigorous reasoning will determine which tasks and business processes will be impacted by their widespread deployment.
Visit Moody’s website to find out how Moody’s is incorporating cutting-edge technologies, such as artificial intelligence, to help banks meet their existing challenges more effectively.
Related Link: BIS Bulletin
Keywords: International, Banking, BIS, RegTech, Generative AI, SupTech
Previous Article
ECB is Conducting First Cyber Risk Stress Test for BanksRelated Articles
OSFI Issues Phase2 Consultation on Climate Scenario Exercise for Banks
The Office of the Superintendent of Financial Institutions (OSFI) recently announced a consultation on the second phase of the Standardized Climate Scenario Exercise (SCSE) for banks and other financial institutions it regulates in Canada.
BIS and Central Banks Experiment with GenAI to Assess Climate Risks
A recent report from the Bank for International Settlements (BIS) Innovation Hub details Project Gaia, a collaboration between the BIS Innovation Hub Eurosystem Center and certain central banks in Europe
Nearly 25% G-SIBs Commit to Adopting TNFD Nature-Related Disclosures
Nature-related risks are increasing in severity and frequency, affecting businesses, capital providers, financial systems, and economies.
Singapore to Mandate Climate Disclosures from FY2025
Singapore recently took a significant step toward turning climate ambition into action, with the introduction of mandatory climate-related disclosures for listed and large non-listed companies
SEC Finalizes Climate-Related Disclosures Rule
The U.S. Securities and Exchange Commission (SEC) has finalized the long-awaited rule that mandates climate-related disclosures for domestic and foreign publicly listed companies in the U.S.
EBA Proposes Standards Related to Standardized Credit Risk Approach
The European Banking Authority (EBA) has been taking significant steps toward implementing the Basel III framework and strengthening the regulatory framework for credit institutions in the EU
US Regulators Release Stress Test Scenarios for Banks
The U.S. regulators recently released baseline and severely adverse scenarios, along with other details, for stress testing the banks in 2024. The relevant U.S. banking regulators are the Federal Reserve Bank (FED), the Federal Deposit Insurance Corporation (FDIC), and the Office of the Comptroller of the Currency (OCC).
Asian Governments Aim for Interoperability in AI Governance Frameworks
The regulatory landscape for artificial intelligence (AI), including the generative kind, is evolving rapidly, with governments and regulators aiming to address the challenges and opportunities presented by this transformative technology.
EBA Proposes Operational Risk Standards Under Final Basel III Package
The European Union (EU) has been working on the final elements of Basel III standards, with endorsement of the Banking Package and the publication of the European Banking Authority (EBA) roadmap on Basel III implementation in December 2023.
EFRAG Proposes XBRL Taxonomy and Standard for Listed SMEs Under ESRS
The European Financial Reporting Advisory Group (EFRAG), which plays a crucial role in shaping corporate reporting standards in European Union (EU), is seeking comments, until May 21, 2024, on the Exposure Draft ESRS for listed SMEs.