Follow Us! ThumbsUp
info@seon.io
False Negatives

What Are False Negatives in Fraud?

False negatives in fraud are where fraudulent transactions or activities are incorrectly identified by anti-fraud systems as legitimate, and therefore allowed to proceed. Every false negative represents an incident where a criminal has managed to bypass the security measures in place.

Unsurprisingly, a false negative is the opposite of a false positive. A false positive is where an anti-fraud system incorrectly identifies a genuine transaction as fraudulent.

In 2022, online payment fraud alone accounted for business losses of around $41 billion. As such, reducing false negatives is a key priority for all companies.

What Is the False Negatives Rate?

The false negatives rate is based on a simple calculation and is usually expressed as a percentage.

For example, if four out of 100 transactions were permitted to proceed despite being fraudulent, and despite the measures in place, this would represent a false negatives rate of 4%.

Alongside the false negatives rate, businesses typically also monitor the following:

  • True negatives: Where legitimate transactions are correctly permitted.
  • False positives: Where genuine transactions are incorrectly flagged as fraudulent.
  • True positives: Where anti-fraud systems do their job correctly and block fraudulent activity.

What Causes False Negatives?

Various things can cause false negatives. It is not possible to eliminate them completely, as fraudsters constantly work to successfully commit their crimes despite the measures in place.

Here are some examples of how false negatives can occur:

1. Inadequate or outdated anti-fraud detection measures.

2. Fraudsters circumventing rudimentary, rules-based anti-fraud systems. For example, they may learn that a fraud detection & prevention tools has a threshold in place to provide especially thorough checks for transactions of over $5,000. The criminals could then decide to only attempt transactions below that amount.

3. Poorly configured machine learning systems incorrectly flagging transactions as genuine.

4. Fraudsters using advanced techniques to avoid detection, such as the use of false identities.

Take Your Fraud Prevention to the Next Level

See how SEON’s powerful fraud detection software can support your business to reduce fraud by up to 70%.

Learn More

Why Are False Negatives a Problem?

False negatives are a serious problem because every incident represents a time when a fraudster has successfully bypassed the anti-fraud measures in place. A high false negatives rate suggests the following:

  • There may be shortcomings in the anti-fraud systems in place, or the system may be poorly configured.
  • The balance between preventing incidents of fraud and avoiding false positives and customer friction may require review.

Reducing the incidence of false negatives is a constant and ongoing priority for all businesses dealing with financial transactions.

How Potentially Damaging Are False Negatives?

The most obvious damage caused by false negatives is the financial loss that can occur from the initial fraudulent transaction(s). This could mean a fraudster obtaining anything from goods to a credit card or loan. However, the potential damage can go far beyond that.

False negatives can cause a company to be in breach of its obligations under anti-money laundering (AML) regulations. Penalties for this increased by over 50% in 2022, with firms fined nearly $5 billion in total.

Once a fraudster has managed to “achieve” a false negative and opened an account, made a purchase, or gained access to a system, they may then be able to make additional transactions or financial gain before the business in question detects them.

Five Ways to Avoid False Negatives

Here are some of the ways that companies can avoid false negatives and reduce their false negatives rate:

  1. Augment fraud detection systems with technologies such as digital footprint analysis and device and browser fingerprinting, ensuring the best chance of successfully flagging fraud attempts.
  2. Establish a comprehensive system of manual reviews to help validate automated decision-making.
  3. Use whitebox machine learning – also known as explainable machine learning. This allows humans to have full visibility of why AI-based systems are making the decisions they are making.
  4. Regularly review the systems in place – with an eye toward the latest fraud techniques and trends.
  5. Proactively tweak the balance between preventing fraud and minimizing customer friction.

Related Terms

Related Articles

Sources

Contact Us for a Demo

Feel free to reach out to us for a demo!