Credit card fraud

Given the immense difficulty of detecting credit card fraud, artificial and computational intelligence was developed in order to make machines attempt tasks in which humans are already doing well. Computation intelligence is simply a subset of AI enabling intelligence in a changing environment. Due to advances in both artificial and computational intelligence, the most commonly used and suggested ways to detect credit card fraud are rule induction techniques, decision trees, neural networks, Support Vector Machines, logistic regression, and meta heuristics. There are many different approaches that may be used to detect credit card fraud. For example, some "suggest a framework which can be applied real time where first an outlier analysis is made separately for each customer using self-organizing maps and then a predictive algorithm is utilized to classify the abnormal looking transactions." Some problems that arise when detecting credit card fraud through computational intelligence is the idea of misclassifications such as false negatives/positives, as well as detecting fraud on a credit card having a larger available limit is much more prominent than detecting a fraud with a smaller available limit. One algorithm that helps detect these sorts of issues is determined as the MBO Algorithm. This is a search technique that brings upon improvement by its "neighbor solutions". Another algorithm that assists with these issues is the GASS algorithm. In GASS, it is a hybrid of genetic algorithms and a scatter search.[8]

Touching a little more on the difficulties of credit card fraud detection, even with more advances in learning and technology every day, companies refuse to share their algorithms and techniques to outsiders. Additionally, fraud transactions are only about 0.01–0.05% of daily transactions, making it even more difficult to spot. Machine learning is similar to artificial intelligence where it is a sub field of AI where statistics is a subdivision of mathematics.  With regards to machine learning, the goal is to find a model that yields that highest level without overfitting at the same time. Overfitting means that the computer system memorized the data and if a new transaction differs in the training set in any way, it will most likely be misclassified, leading to an irritated cardholder or a victim of fraud that was not detected. The most popular programming used in machine learning are Python, R, and MatLab. At the same time, SAS is becoming an increasing competitor as well. Through these programs, the easiest method used in this industry is the Support Vector Machine. R has a package with the SVM function already programmed into it. When Support Vector Machines are employed, it is an efficient way to extract data. SVM is considered active research and successfully solves classification issues as well. Playing a major role in machine learning, it has "excellent generalization performance in a wide range of learning problems, such as handwritten digit recognition, classification of web pages and face detection." SVM is also a successful method because it lowers the possibility of overfitting and dimensionality.[9]

Application fraud takes place when a person uses stolen or fake documents to open an account in another person's name. Criminals may steal or fake documents such as utility bills and bank statements to build up a personal profile. When an account is opened using fake or stolen documents, the fraudster could then withdraw cash or obtain credit in the victim's name.[10]

Application fraud can also occur using a synthetic identity which is similar to the fake documents mentioned above. A synthetic identity is personal information gathered from many different identities to create one fake identity.[11] Once the identity and the account is established, the fraudster has a few different options to take advantage of the bank.  They can maximize their credit card spending by spending as much money as possible on their new credit card. Many fraudsters will use the new credit card to purchase items that have a high resale value so they can turn it into cash.

An account takeover refers to the act by which fraudsters will attempt to assume control of a customer's account (i.e. credit cards, email, banks, SIM card and more). Control at the account level offers high returns for fraudsters. According to Forrester, risk-based authentication (RBA) plays a key role in risk mitigation.[12]

A fraudster uses parts of the victim's identity such as an email address to gain access to financial accounts. This individual then intercepts communication about the account to keep the victim blind to any threats. Victims are often the first to detect account takeover when they discover charges on monthly statements they did not authorize or multiple questionable withdrawals.[13] There has been an increase in the number of account takeovers since the adoption of EMV technology, which makes it more difficult for fraudsters to clone physical credit cards.[14]

Among some of the most common methods by which a fraudster will commit an account, takeover includes proxy-based "checker" one-click apps, brute-force botnet attacks, phishing,[15] and malware. Other methods include dumpster diving to find personal information in discarded mail, and outright buying lists of 'Fullz', a slang term for full packages of identifying information sold on the black market.[16]

Once logged in, fraudsters have access to the account and can make purchases and withdraw money from bank accounts.[17] They have access to any information that is tied to the account, they can steal credit card numbers along with social security numbers. They can change the passwords to prevent the victim from accessing their account. Cybercriminals have the opportunity to open other accounts, utilize rewards and benefits from the account, and sell this information to other hackers.

Social engineering fraud can occur when a criminal poses as someone else which results in a voluntary transfer of money or information to the fraudster. Fraudsters are turning to more sophisticated methods of scamming people and businesses out of money. A common tactic is sending spoof emails impersonating a senior member of staff and trying to deceive employees into transferring money to a fraudulent bank account.[18]

Fraudsters may use a variety of techniques in order to solicit personal information by pretending to be a bank or payment processor. Telephone phishing is the most common social engineering technique to gain the trust of the victim.

Businesses can protect themselves with a dual authorisation process for the transfer of funds that requires authorisation from at least two persons, and a call-back procedure to a previously established contact number, rather than any contact information included with the payment request. The bank must refund any unauthorised payment; however, they can refuse a refund if they can prove the customer authorised the transaction, or it can prove the customer is at fault because they acted deliberately, or failed to protect details that allowed the transaction.[19]

Green plastic unit on an ATM slot, intended to stop thieves from installing a skimmer device on the machine

Skimming is the theft of personal information which has been used in an otherwise normal transaction. The thief can procure a victim's card number using basic methods such as photocopying receipts or more advanced methods such as using a small electronic device (skimmer) to swipe and store hundreds of victims' card numbers. Common scenarios for skimming are taxis, restaurants or bars where the skimmer has possession of the victim's payment card out of their immediate view.[20] The thief may also use a small keypad to unobtrusively transcribe the three or four-digit card security code, which is not present on the magnetic strip.

Call centers are another area where skimming can easily occur.[21] Skimming can also occur at merchants when a third-party card-reading device is installed either outside a card-swiping terminal. This device allows a thief to capture a customer's card information, including their PIN, with each card swipe.[22]

Skimming is difficult for the typical cardholder to detect, but given a large enough sample, it is fairly easy for the card issuer to detect. The issuer collects a list of all the cardholders who have complained about fraudulent transactions, and then uses data mining to discover relationships among them and the merchants they use. Sophisticated algorithms can also search for patterns of fraud. Merchants must ensure the physical security of their terminals, and penalties for merchants can be severe if they are compromised, ranging from large fines by the issuer to complete exclusion from the system, which can be a death blow to businesses such as restaurants where credit card transactions are the norm.

Instances of skimming have been reported where the perpetrator has put over the card slot of an automated teller machine, a device that reads the magnetic strip as the user unknowingly passes their card through it.[23] These devices are often used in conjunction with a miniature camera to read the user's personal identification number at the same time.[24] This method is being used in many parts of the world, including South America, Argentina,[25] and Europe.[26]

Online bill paying or internet purchases utilizing a bank account are a source for repeat billing known as "recurring bank charges". These are standing orders or banker's orders from a customer to honour and pay a certain amount every month to the payee. With E-commerce, especially in the United States, a vendor or payee can receive payment by direct debit through the ACH Network. While many payments or purchases are valid, and the customer has intentions to pay the bill monthly, some are known as Rogue Automatic Payments.[27]

Another type of credit card fraud targets utility customers. Customers receive unsolicited in-person, telephone, or electronic communication from individuals claiming to be representatives of utility companies. The scammers alert customers that their utilities will be disconnected unless an immediate payment is made, usually involving the use of a reloadable debit card to receive payment. Sometimes the scammers use authentic-looking phone numbers and graphics to deceive victims.

Phishing is one of the most common methods used to steal personal data. It is a type of cyber attack in which the attacker acts as a credible person, institution, or entity and attempts to lure the victim into accepting a message or taking action with the specific request. Often, the target of the attack will receive an email or text message about something they would possibly want or need with the hope of tricking them into opening or downloading the message. During the COVID-19 pandemic, phishing has been on the rise as our world turned even more virtual. To give perspective, "researchers noted a substantial spike of 667% in COVID-19 phishing attacks in the first months of the pandemic."[28]. Also, given the significance of health care systems over these recent years health care companies have been the main targets of phishing attacks. These companies have tons of personal data stored that can be extremely valuable to the attacker.

Information sharing is the transfer or exchange of data between individuals, companies, organizations, and technologies. Advances in technology, the internet, and networks have accelerated the growth of information sharing. Information is spread and shared in the matter of seconds, and is being accumulated and digested at speeds faster than ever before. People are often not aware of how much sensitive and personal information they share every day. For example, when purchasing goods online, the buyer's name, email address, home address, and credit card information are stored and shared with third parties to track them and their future purchases. Organizations work hard to keep individuals' personal information secure in their databases, but sometimes hackers are able to compromise its security and gain access to an immense amount of data. One of the largest data breaches occurred at the discount retailer Target. In this breach about 40 million shopper were affected. In this specific case, the hackers targeted their point-of-sale system – meaning "they either slipped malware into the terminals where customers swipe their credit cards, or they collected customer data while it was on route from Target to its credit card processors."[29] In just one single purchase at the register, masses of personal data is collected which when stolen has major ramifications. The financial infrastructure and payment system will continue to be a work-in-progress as it constantly is at battle with security hackers.

While not federally mandated in the United States PCI DSS is mandated by the Payment Card Industry Security Standard Council, which is composed of major credit card brands and maintains this as an industry standard. Some states have incorporated the standard into their laws.

The US Department of Justice announced in September 2014 that it will seek to impose a tougher law to combat overseas credit card trafficking. Authorities say the current statute is too weak because it allows people in other countries to avoid prosecution if they stay outside the United States when buying and selling the data and do not pass their illicit business through the U.S. The Department of Justice asks US Congress to amend the current law that would make it illegal for an international criminal to possess, buy or sell a stolen credit card issued by a U.S. bank independent of geographic location.[30]

In the US, federal law limits the liability of cardholders to $50 in the event of theft of the actual credit card, regardless of the amount charged on the card, if reported within 60 days of receiving the statement.[31] In practice, many issuers will waive this small payment and simply remove the fraudulent charges from the customer's account if the customer signs an affidavit confirming that the charges are indeed fraudulent. If the physical card is not lost or stolen, but rather just the credit card account number itself is stolen, then federal law guarantees cardholders have zero liability to the credit card issuer.[32]

In the UK, credit cards are regulated by the Consumer Credit Act 1974 (amended 2006). This provides a number of protections and requirements. Any misuse of the card, unless deliberately criminal on the part of the cardholder, must be refunded by the merchant or card issuer.

The regulation of banks in the United Kingdom is undertaken by the: Bank of England (BoE); Prudential Regulation Authority (PRA) a division of the BoE; and the Financial Conduct Authority (FCA) who manages the day to day oversight. There is no specific legislation or regulation that governs the credit card industry. However, the Association for Payment Clearing Services (APACS) is the institution that all settlement members are a part of. The organisation works under the Banking Consolidation Directive to provide a means by which transactions can be monitored and regulated.[33] UK Finance is the association for the UK banking and financial services sector, representing more than 250 firms providing credit, banking and payment-related services.

A graph showing the number of victims and proportion of the population or household affected by different offences

In Australia, credit card fraud is considered a form of identity crime. The Australian Transaction Reports and Analysis Centre has established standard definitions in relation to identity crime for use by law enforcement across Australia:

• The term identity encompasses the identity of natural persons (living or deceased) and the identity of bodies corporate
• Identity fabrication describes the creation of a fictitious identity
• Identity manipulation describes the alteration of one's own identity
• Identity theft describes the theft or assumption of a pre-existing identity (or significant part thereof), with or without consent and whether, in the case of an individual, the person is living or deceased
• Identity crime is a generic term to describe activities/offences in which a perpetrator uses a fabricated identity, a manipulated identity, or a stolen/assumed identity to facilitate the commission of a crime(s).[34]

Given increasing number of unauthorised payment card transactions involving frauds and scams, the Hong Kong Monetary Authority issued two Circulars on 25 April 2023. [35]

Estimates created by the Attorney-General's Department show that identity crime costs Australia upwards of $1.6 billion each year, with the majority of about $900 million being lost by individuals through credit card fraud, identity theft and scams.[34] In 2015, the Minister for Justice and Minister Assisting the Prime Minister for Counter-Terrorism, Michael Keenan, released the report Identity Crime and Misuse in Australia 2013–14. This report estimated that the total direct and indirect cost of identity crime was closer to $2 billion, which includes the direct and indirect losses experienced by government agencies and individuals, and the cost of identity crimes recorded by police.[36]

The victim of credit card fraud in Australia, still in possession of the card, is not responsible for anything bought on it without their permission. However, this is subject to the terms and conditions of the account. If the card has been reported physically stolen or lost the cardholder is usually not responsible for any transactions not made by them, unless it can be shown that the cardholder acted dishonestly or without reasonable care.[34]

• PAN truncation – not displaying the full primary account number on receipts
• Tokenization (data security) – using a reference (token) to the card number rather than the real card number
• Requesting additional information, such as a PIN, ZIP code, or Card Security Code
• Performing geolocation validation, such as IP address
• Use of Reliance Authentication, indirectly via PayPal, or directly via iSignthis or miiCard.

• Fraud detection and prevention software[49][50][51][52] that analyzes patterns of normal and unusual behavior as well as individual transactions in order to flag likely fraud. Profiles include such information as IP address.[53] Technologies have existed since the early 1990s to detect potential fraud. One early market entrant was Falcon;[50] other leading software solutions for card fraud include Actimize, SAS, BAE Systems Detica, and IBM.
• Fraud detection and response business processes such as:
  • Contacting the cardholder to request verification
  • Placing preventative controls/holds on accounts that may have been victimized
  • Blocking card until transactions are verified by the cardholder
  • Investigating fraudulent activity
• Strong Authentication measures such as:
  • Multi-factor Authentication, verifying that the account is being accessed by the cardholder through requirement of additional information such as account number, PIN, ZIP, challenge questions. There are five main factors to multi-factor authentication and they include:[54]
    1. Knowledge - things a user knows such as passwords or answers to secret questions.
    2. Possession - an object the user should have in their possession such as the actual credit card.
    3. Inherence - a biological trait of the user such as finger-print or facial recognition.
    4. Location - where the user is at the time of the authentication - verify the user was the one to use the card.
    5. Time - when the authentication is taking place - is it a strange hour or multiple times?
  • Multi possession-factor authentication, verifying that the account is being accessed by the cardholder through requirement of additional personal devices such as smart watch, smart phone challenge–response authentication[55]
  • Out-of-band Authentication,[56] verifying that the transaction is being done by the cardholder through a "known" or "trusted" communication channel such as text message, phone call, or security token device
• Industry collaboration and information sharing about known fraudsters and emerging threat vectors[57][58]

• Automated Data Controls:

1. The use of automated data controls which are used to recognize when unusual activity or spending occur with a credit card. These controls can be used in real time to react "...to anything suspicious they come upon, so the flow of fraudulent activity is stopped as soon as possible..." (Johnston).[59] The three main ways automated data controls protect information includes:
   1. Reconciliation and verification to ensure that the controls are working properly.
   2. Continuous monitoring and alerting which alerts the cardholder/bank when unusual activity is taking place.
   3. Reporting which ensures organizations have proper controls in place to prevent fraudulent activity

• Internal self-banking area for the customer to carry out the transactions regardless of the weather conditions. The access door:
  • Identifies every cardholder that gains access to the designated area
  • Increases protection for customers during self-service procedures
  • Protects the ATMs and banking assets against unauthorized usage
  • The protected area can also be monitored by the bank's CCTV system
  • Cards use CHIP identification (ex PASSCHIP [60]) to decrease the possibility of card skimming

• Enacting consumer protection laws related to card fraud
• Performing regular examinations and risk assessments of credit card issuers[61]
• Publishing standards, guidance, and guidelines for protecting cardholder information and monitoring for fraudulent activity[62]
• Regulation, such as that introduced in the SEPA and EU28 by the European Central Bank's 'SecuRe Pay'[63] requirements and the Payment Services Directive 2[64] legislation.

• Reporting lost or stolen cards
• Reviewing charges regularly and reporting unauthorized transactions immediately
• Keeping a credit card within the cardholder's view at all times, such as in restaurants and taxis
• Installing virus protection software on personal computers
• Using caution when using credit cards for online purchases, especially on non-trusted websites, make sure site is reputable
• Keeping a record of account numbers, their expiration dates, and the phone number and address of each company in a secure place.[65]
• Not sending credit card information by unencrypted email
• Not keeping written PIN numbers with the credit card.
• Not giving out credit card numbers and other information online
• Sign up for transaction alerts when card is used[66]
• Be aware of phishing schemes