In this paper we evaluate the capabilities of LLM Agents in generating code for real-world problems. Specifically, we explore code synthesis for microservice-based applications, a widely used architectural pattern for building applications. We define a standard template for specifying these applications, and we propose a metric for scoring the difficulty of a specification. The higher the score, the more difficult it is to generate code for the specification. Our experimental results show that agents using strong LLMs (like GPT-3o-mini) do fairly well on medium difficulty specifications but do poorly on those of higher difficulty levels. This is due to more intricate business logic, a greater use of external services, database integration and inclusion of non-functional capabilities such as authentication. We analyzed the errors in LLM-synthesized code and report on the key challenges LLM Agents face in generating code for these specifications. Finally, we show that using a fine-grained approach to code generation improves the correctness of the generated code.

Ensuring fairness in transaction fraud detection models is vital due to the potential harms and legal implications of biased decision-making. Despite extensive research on algorithmic fairness, there is a notable gap in the study of bias in fraud detection models, mainly due to the field's unique challenges. These challenges include the need for fairness metrics that account for fraud data's imbalanced nature and the tradeoff between fraud protection and service quality. To address this gap, we present a comprehensive fairness evaluation of transaction fraud models using public synthetic datasets, marking the first algorithmic bias audit in this domain. Our findings reveal three critical insights: (1) Certain fairness metrics expose significant bias only after normalization, highlighting the impact of class imbalance. (2) Bias is significant in both service quality-related parity metrics and fraud protection-related parity metrics. (3) The fairness through unawareness approach, which involved removing sensitive attributes such as gender, does not improve bias mitigation within these datasets, likely due to the presence of correlated proxies. We also discuss socio-technical fairness-related challenges in transaction fraud models. These insights underscore the need for a nuanced approach to fairness in fraud detection, balancing protection and service quality, and moving beyond simple bias mitigation strategies. Future work must focus on refining fairness metrics and developing methods tailored to the unique complexities of the transaction fraud domain.

Ryan Owen holds an MBA from the University of South Carolina, and has rich experience in financial services, having worked with Liberty Mutual, Sun Life, and other financial firms. Ryan writes and edits AI industry trends and use-cases for Emerj's editorial and client content. American Express began as a freight forwarding company in the mid-19th century. Expanding over time to include financial products and travel services, American Express today reports some 114 million cards in force and $1.2 trillion in billed business worldwide. American Express trades on the NYSE with a market cap that exceeds $136 billion, as of November 2021.

Thanks to the internet, we no longer need to go to the shops; instead, the shops come to us. In a few clicks you can order everything from the latest digital gadgets to dog food, from the comfort of your sofa. And same-day delivery options mean you can receive items faster than ever. But the speedy online transactions and one-click purchasing systems that underpin the ecommerce sector don't just make life easier for consumers; they make things easier for fraudsters too. Successful ecommerce retailers receive thousands of orders a day, and these card-not-present (CNP) purchases are harder to verify than those where the card and cardholder are physically present.

Man-made brainpower (AI) will soon be at the core of each major technological framework on the planet to manage and get to your strategic information. Only a couple of uses are cyber and homeland security, anti-money laundering, payments, financial markets, biotech, healthcare, marketing, natural language processing (NLP), computer vision, electrical grids, nuclear power plants, air traffic control, and Internet of Things (IoT). Artificial Intelligence is turning into a significant staple of innovation, scarcely any individuals comprehend the advantages and weaknesses of AI and Machine Learning innovations. While machine intelligence is sure to assume a key role in the making of cutting edge frameworks in a wide assortment of industry areas sooner rather than later, it is especially applicable in quickly developing businesses, for example, ICT, manufacturing and transportation. Over the globe, mobile operators are preparing to deploy the fifth era of 3GPP mobile wireless networks (5G).

When it comes to using artificial intelligence (AI) for credit-card fraud management, the benefits are more than meets the eye. Yes, AI-powered detection systems spot fraud more quickly and help contain losses. But the systems also put a new spin on fraud for credit card providers: Real-time fraud detection helps inspire customer confidence and boost loyalty. Call it the upside of fraud. According to at least one study, organizations with a formalized program in place to monitor and prevent fraud dramatically improve their customer satisfaction and retention rates.

These great deals are the perfect way to jumpstart your weekend. If you make a purchase by clicking one of our links, we may earn a small share of the revenue. However, our picks and opinions are independent from USA Today's newsroom and any business incentives. There are so many reasons to get pumped that the weekend has finally arrived. All the great deals you can grab on everything from the latest fashion and beauty products to meat thermometers and more.

By using artificial intelligence (AI), Visa Inc. helped issuers prevent an estimated $25 billion in annual fraud, the company announced on June 17. The company accomplished this using Visa Advanced Authorization (VAA), a comprehensive risk management tool that monitors transaction authorization on the Visa global network, VisaNet, in real time. VAA evaluates every single transaction on VisaNet and helps issuers swiftly identify emerging fraud trends and patterns, allowing the issuers to respond promptly to instances of fraud, while approving legitimate transactions. "One of the toughest challenges in payments is separating good transactions made by cardholders from bad ones attempted by fraudsters without adding friction to the process," said Melissa McSherry, senior vice president and global head of Data Products and Solutions at Visa. The speed with which Visa can evaluate a transaction is crucial.

Visa has rolled out internal data that indicates its artificial intelligence (AI)-based Advanced Authorization (VAA) security product has helped issuers prevent an estimated $25 billion in annual fraud. VAA is a risk management tool that monitors and evaluates transactions over VisaNet in real time to helps issuer "see" fraud as it happens and shut it down based on its ability to spot emerging fraud patterns and trends. Over 127 billion transactions flowed across VisaNet in 2018 between merchants and financial institutions on VisaNet last year, and AI was used to analyze 100 percent of those transactions. Each bit of analysis and fraud ranking takes about one millisecond -- so financial institutions can approve legitimate and bounce bad ones without the customer ever feeling a delay. "One of the toughest challenges in payments is separating good transactions made by cardholders from bad ones attempted by fraudsters without adding friction to the process," said Visa Senior Vice president and Global Head of Data Products and Solutions Melissa McSherry.

Machine learning and data mining techniques have been used extensively in order to detect credit card frauds. However, most studies consider credit card transactions as isolated events and not as a sequence of transactions. In this article, we model a sequence of credit card transactions from three different perspectives, namely (i) does the sequence contain a Fraud? (ii) Is the sequence obtained by fixing the card-holder or the payment terminal? (iii) Is it a sequence of spent amount or of elapsed time between the current and previous transactions? Combinations of the three binary perspectives give eight sets of sequences from the (training) set of transactions. Each one of these sets is modelled with a Hidden Markov Model (HMM). Each HMM associates a likelihood to a transaction given its sequence of previous transactions. These likelihoods are used as additional features in a Random Forest classifier for fraud detection. This multiple perspectives HMM-based approach enables an automatic feature engineering in order to model the sequential properties of the dataset with respect to the classification task. This strategy allows for a 15% increase in the precision-recall AUC compared to the state of the art feature engineering strategy for credit card fraud detection.