AI Shield

AI SHIELD

Reduce direct losses to fraud with our real-time AI powered anti-fraud tool – get 97% accuracy in fraud detection and prevention thanks to ML algorithms and Big Data analysis.

Protect your voice traffic with the power of Machine Learning

Tilda Publishing

Detect fraud in real-time with unparalleled accuracy and simplicity

AI Shield was designed to solve the technological and business problems caused by fraud. It allows your company to:

Save time and money from Day 1

Our engine is trained on vast volumes of retail and wholesale voice traffic, providing unparalleled accuracy in fraud detection out-of-the-box.

Increase revenue by avoiding service denial

Enjoy instant flexibility with a user-friendly interface

Customize alerting, blocking, and reporting policies in a click and reduce operational costs.

Our built-in IVR module warns your enterprise customers when they are dialing high-risk number ranges.

See AI Shield in action

AI Shield uses multiple levels of protection that interact in real time
to provide highly accurate fraud detection.

AI Shield is a cloud-based FMS that detects and blocks all major types of voice fraud attacks. It is a fully-tested anti-fraud tool for telcos who want to protect their networks from fraudulent activity and eliminate direct, operational, and reputational losses.

Multi-level, proactive monitoring and a range of key features offer a stunning level of security and ease-of-use. AI Shield is an effortless and robust way to keep your network and subscribers protected against fraudulent interference.

Enjoy the all-in-one interface

•‎ Monitor the fraud landscape in real time – see all parameters of fraudulent calls on one dashboard
•‎ Reverse active blockages in one click
•‎ Store and edit your blacklists, whitelists, and threshold-based rules

With our user-friendly interface it's simple to:

Prevent
well-disguised attacks

•‎ Wangiri & Wangiri 2.0
•‎ PBX hack
•‎ Short stopping
•‎ Robocalls and Spam calls
•‎ Arbitrage

AI Shield protects operators and enterprises from major types of fraud:

If you suffer from CLI spoofing, Call stretching or SIM box Fraud, the AB Handshake Call Validation module will help.

Be in control with simple and secure integration

AI SHIELD can be installed inside the security perimeter of an operator and in the Cloud. It is connected to any network element that can transmit A and B numbers and 3 call events (start, connect, stop). This means that our integration options are quick, secure, and flexible.

Following call event analysis by the AI engine, the connector sends call control messages to the operator's network node, if required by the blocking policy.

Explore our reputable association memberships

Our membership in the largest communities in the telecom industry is a testament to the excellence and effectiveness of AI Shield technology.

Member of the CFCA Trusted Calls Task Force Committee, aimed at developing the most effective way to combat communications fraud

CFCA

Member of the GSMA, a member-led global organization shaping the future of the connectivity ecosystem

GSMA

Tilda Publishing

FAQ

How much time is required for AI training after integration?

The tool already comes trained on a large number of calls. We guarantee protection from Day 1.

Is the system updated by AB Handshake after integration?

Yes, we perform regular updates for standalone installations so that clients can benefit from cases detected in other networks (in an anonymized format).

What is the accuracy of the tool?

The False Discovery Rate (FDR) of AI Shield is <3% and the False Positive Rate is <0.001% which makes it a very accurate and precise tool for fraud prevention and detection.

The False Discovery Rate (FDR) is calculated using the following formula:
FDR = FP/(FP+TP)
Where:
• FP - False Positives
• TP - True Positives

The False Positive Rate (FPR) is calculated using the following formula:
FPR= FP/(FP+TN)
Where:
• FP - False Positives
• TN - True Negatives