Category: Fraud Management

The telecoms industry has always suffered from the malicious impacts of fraudsters. In this digital era, traditional telecom services are replaced with new services, giving rise to new fraud types. Therefore, technology to support fraud management operations has changed significantly, with an increased reliance on Artificial Intelligence and Machine Learning as a means for uncovering and identifying fraud. In one of its recent reports, Gartner estimated that by 2022, new implementations of ML within CSP fraud management would reduce fraud losses by 10%.

Will AI/ML be the future for Fraud detection & prevention?

To help us understand this further, it gives us immense pleasure to have interviewed Joseph Nderitu

How has the telco fraud landscape changed in recent years?

There has been significant evolution into new technologies, different access methods, and the proliferation of IP technologies. This means the fraudster may be seated inside the telco premises or half-way around the world and has the flexibility to “hit” the telco in any number of ways. For operators who are offering mobile money, they have suddenly found themselves running a bank on top of their IP network, something that was not in their corporate DNA, so to speak. To cap it all, as we move into the 5G and IoT landscape, the volume of data will be very high whilst the response time to fraud and leakages will be expected to be faster.

Why do telcos need to continuously innovate and invest in newer technologies to address the risks?

The pressure to address risks is higher than ever. Customer expectations are high, and with rising competition, failure to address risk means a reduction in market share. Shareholder value thus gets eroded if risks are not well managed. Regulators are also keeping a keen eye on service providers. This essentially means the challenges are increasing on the one hand, and the service providers’ responsiveness is expected to increase amidst reduced budgets for tools. Innovation is the only way out. We need better tools and methods, which help us to manage risks better.

Do you think AI/ML is the way forward for telco fraud management? And why?

Certainly – if you look at telcos, they are a goldmine of data, always have been. The trends and patterns are there, but we have not been doing a great job of detecting them because the human capacity to see trends is limited. AI/ML technologies remove that limitation – or, at the very least, reduce is significantly. Thus, we are at a sweet spot of sorts – a confluence of rich data, superior ways of looking at the data, and telco businesses that are hungry for insights that will help them manage risks better. As we place everything on IP technologies, it also means we are at a point where the greying of lines between revenue assurance, fraud management, and cybersecurity is happening much faster. The interplay between these areas is a good area for AI/ML.

What are some of the critical fraud areas, which can be covered to a larger extent with AI/ML?

There is a pattern to each fraud. The fact that we do not always do a good job of spotting it with our traditional systems (which are very much rule-based) does not mean the pattern is not there. With that in mind, it means one can apply supervised and unsupervised learning in a lot of areas. There are operators, in Africa, for example, deploying basic learning models on mobile money frauds and using them to identify commission frauds perpetrated in the distribution chain. The features used in such a model would combine the pattern of records obtained from both the GSM systems and the mobile money systems – something that was quite unwieldy to do using conventional FMS. Thus, there is no problem that is too small or too big to be addressed in this new mode of fraud management. 

Likewise, with IoT, smarter devices do not mean less fraud, for example – it might just mean an expanded landscape where we (operators and customers) can be hit from any angle.

Does the introduction of AI/ML mean reducing team size, more focus on other areas, and complementing the team with other teams?

There will always be a need for warm bodies. Proper use of AI/ML means we can free human beings to do things that machines still cannot do, or at least cannot do so well. For example, resolving issues still calls for people to talk to each other, negotiate, agree, implement, and track progress. There will always be room for that, so in a sense, and we just need to let the machines do what machines do best and leave humans to do what only humans can do(for now).

While there is a need, are telcos really investing in these technologies?

As always, some will move faster than others. However, regardless of the speed of implementation (which is a function of budgets, organizational politics, and other organizational peculiarities), by and large, this is the direction the telcos are taking. Some are approaching vendors to update tools; others are doing their in-house experiments using freely available resources such as python. All in all, I doubt there is anybody who is sitting with their arms crossed, hoping that the advent of AI/ML is a fad that will pass. The technology is and will keep on evolving, whether we keep up or not.

On-demand Webinar: AI Master Class for Fraud Management

Watch Now!

Telecom fraud is increasing at an alarming rate globally and is one of the most significant sources of revenue erosion for every telecom operator. With bottom lines under tremendous pressure, it is now time for CSPs to invest in advanced technologies to strengthen their fraud management systems. This would enable them to stay ahead of fraudsters and prevent frauds before they occur, and reduce the impact on revenues. In its report on “Market Trends: Maximizing the Value of Analytics, Artificial Intelligence and Automation in CSP Fraud Management,” Gartner estimated that by 2022, new implementations of ML within CSP fraud management would reduce fraud losses by 10%.

If a Fraud Management team is planning to integrate AI/ML capabilities into their FMS and is considering rolling out an RFP, here are the 10 key points they need to consider while drafting an RFP for AI/ML-based fraud detection service provider.

1. Assessment of Maturity Level 

Approach to fraud detection using AI/ML technologies could significantly transform from the current controls that CSP might be currently doing. Hence, it would be good to conduct a maturity assessment to understand the maturity of their existing fraud management function and how bringing AI/ML capabilities can enhance this maturity level and lead to more automation.

2. In-house or Outsource  

This might be a critical decision that CSP’s might have to take early in the game to go with in-house data scientists or external platforms and vendors. It’s a fact that many of the AI/ML techniques are available in open source worlds, and internal teams can be used to bootstrap and try out various techniques. However, various aspects need to be taken into account before a decision could be made; some of the critical question would be

• What is the right use case?
• How will it scale to meet the increasing data load requirements?
• What algorithms and hyperparameters to be used for solving the business problem?
• What data sets to be used to build the model?
• What should be the best practices in deploying the model in production?

It is a known fact that acquiring clean data engineering itself might take considerable effort in model building. CSPs should explore data engineering and feature store platforms that can cohesively stitch different facets to enable the organization to bring out the required AI transformation.

3. Auto adaptive learning for AI/ML Models  

CSPs live in a world of changing data patterns. A data pattern that was learned last month might not be valid this month. Hence it is essential to keep in mind how these changing data patterns can be quickly fed to ML algorithms. Auto adaptive learning systems are a type of online or feedback learning system that can adjust to changing patterns and provide better predictions in the long term.  It is essential to have such a framework to re-engineer the data features or re-learn the data trends as and when the business scenarios change.

4. Desired Architecture of AI/ML Model 

While deciding upon the AI/ML model, it is recommended to choose a flexible architecture that is scalable horizontally and vertically to adapt to the changing business load scenarios.  Increasingly CSPs are adopting cloud architectures with Kubernetes backed nodes to do processing. Such architecture will ensure planet-scale reliability as well as on-demand elastic scalability if business use cases demand.

5. Metrics for measuring success 

CSPs must have clear metrics to measure the success of the AI/ML models combating fraud. Some of the key metrics that need to be measured are:

• The accuracy of the AI/ML model
• The efficiency of the AI/ML model
• Coverage provided by the AI/ML model
• Finding the unknown unknowns

To achieve the above, fraud management teams need to ensure that the models are trained and tuned accordingly.

6. Pilot use case 

CSPs need to prioritize which fraud types they need to tackle first. First, it would be ideal to choose a use case that is highly prevalent in the existing business scenario rather than the one with a very low probability of occurrence. It is also important to note that while considering this use case, the model should have enough past historical data to learn from to generate accurate results. Considering such a use case would help assess the efficiency of the model compared to how a traditional FMS model would respond to a similar scenario.

7. Open-source community integration 

With a rise in the number of open source communities, various open-source algorithms are readily available along with their libraries, which can be adopted for the model that is being built. It is critical to then fine-tune the micro characteristics of these algorithms to meet the business requirements. Readily adapting the models can help CSPs reduce the time that needs to be invested in building the model and be upscaled easily.

8. Support of different models 

One of the key aspects to cover while drafting the RFP is to look at what kind of machine learning models can be supported. To start with, CSPs can consider a supervised machine learning model. Further down the line, they can move into exploring the unsupervised, semi-supervised, and deep learning models. Deep learning models can effectively deal with image processing business requirements, such as Contracts monitoring or P.O.S fraud detection.

9. Latency and scalability 

Considering a large amount of data CSPs have access to, it is recommended to have low latency and highly scalable models to deal with the large datasets much faster without consuming ever-growing amounts of resources like memory. Also, it is essential to be mindful that the impact of latency on the business scenario can be different for different use cases.

10. Roadmap & Vision

It is suggested that CSPs should have a roadmap and vision in place for achieving the next 3-5 use cases aligned with their business goals. This needs to be done with a specific time frame in mind. The automation procedures would also free up critical analyst resources that are currently stuck up in mundane tasks, which could increase the coverage of the fraud management function to newer digital products.

The above factors are an indicative list, highlighting the key considerations while drafting an RFP. If CSPs wish to augment their fraud management strategy and are looking to roll out an RFP that aligns with the modern-day needs of a Proactive Fraud Management Solution that can optimize and accelerate the digitalization path, now is the time to act. This will enable CSPs to fight against fraud by providing adequate protection that is needed to tackle issues upfront in a preventative manner to avoid loss impacts.

Planning to roll out an RFP any time soon? What are your thoughts about the above points?

Let us know here

Amid the fierce competition that the telecom industry is facing, sometimes we listen to stories of how lack of forethought of one Telco brings on illegal traffic on the network. This has serious implications such as revenue loss and unwarranted accusations among the operators affected by the fraud.

Interconnect Bypass fraud is one such telecom scam costing the industry several billion dollars every year. It brings collateral damage to the networks involved, along with a massive impact on their profitability. Telcos could be imposed a hefty penalty for their failure to detect and resolve the issue on time. Further, it could bring serious business implications for all participating telcos. In the process of rampant blocking of suspicious traffic, sometimes traffic of genuine customers could get blocked, leading to customer dissonance and dissatisfaction along with loss of other business opportunities.

Here’s an example of a Telco who suffered massively due to Bypass fraud.

Why did this happen?

The Telco had been massively impacted by off-net Bypass fraud, where the network of the operator was being misused to land fraudulent calls on the competitor’s system. Over time, the problem became so grave that the Regulatory Authority of the country had to step in and take charge of things. This eventually ended with the competitors blocking both fraudulent and genuine traffic from the Telco affected by the interconnection fraud.

Investigations conducted confirmed that the huge differences between the International and local termination rates made the environment suitable for fraudsters to run their schemes. Moreover, it was also identified that there were not enough KYC controls in the country to facilitate certain onboarding checks which distinguish a genuine customer from a fraudulent one.

The Repercussions

There were multiple warnings and memos issued to the operator from the Regulator, indicating that the operator would have to face penalties if amendments are not made in time.

Customers flooded the operator with complaints saying that their off-net calls were being barred without prior notice and for no fault of theirs. They also threatened that they would eventually churn out of the network if their services weren’t restored.

The atmosphere grew so tense that instead of cooperating, the operators became more aggressive and indulged in a rat-race in trying to prove a point to the Regulator as to how better and efficient they were from the rivals in terms of detecting Bypass fraud cases.

The solution

With the understanding that Bypass frauds are rampant, Telcos need to direct their efforts towards incorporating advanced technologies such as AI/ML in their fraud management systems. This would help Telcos with early detection of fraud, as quickly as within 10 minutes of usage. Also, this would enable telcos to identify the various hidden complex patterns that otherwise might go undetected using traditional rule-based methods. These technologies can also help in monitoring the changing behavior without manual intervention.

Furthermore, telcos need to build cognizance in terms of fraudulent behavior and the locations from where they are generated. Telcos need to understand what kind of products tend to get misused by these fraudsters. This would help telcos to nip things in the bud thus saving millions and keeping their brand image intact.

On-Demand Webinar: Addressing Bypass Fraud through Artificial Intelligence

Watch Now

Africa owes much of its recent economic growth to the use of telecommunications services. However, over the past few years, telcos in the region have been hit by several telecom frauds, most notably SIM Box fraud, also known as the interconnect bypass fraud. The unfortunate reality across the region is that SIM Box fraud continuous to be one of the major frauds affecting the dynamic telecom market in Africa. Despite the best efforts taken by telcos to curb this issue, the impact from this fraud is huge in terms of the loss in revenues to telcos and taxes to the regulatory.

Let us look at a couple of recent instances,  to put this scenario into context. Recently, the Ugandan Communications Commission (UCC), in collaboration with the local police, conducted an anti-telecom fraud operation that resulted in the seizure of over 1,500 SIM cards and 4 SIM Boxes.  In another incident, the National Communications Authority (NCA), the Ghana Revenue Authority (GRA), KelniGVG, and the Ghana Police Service retrieved 11 SIM Box equipment, each capable of using 352 SIM cards simultaneously from Tema, Ghana. Suspected criminals caused a massive loss of about $2mn per month.

The African market is at a period where telcos are struggling to find a solution towards shrinking margins while being tasked to address the ever-increasing customer demands by investing in new technologies. During such a time the threat posed by SIM Box fraud will impede the growth of mobile telephony across the continent, which will further hamper the region’s journey towards digital transformation.

Why SIM Box Fraudsters still Target Africa? 

• As per the industry reports, the African mobile market in the region will reach several important milestones over the next five years: half a billion mobile subscribers in 2021, 1 billion mobile connections in 2024, and 50% subscriber penetration by 2025. Smartphone adoption will continue to rise rapidly in the region, reaching 50% of total connections by the end of 2020, as cheaper devices have become available. Fraudsters operating SIM Boxes are taking advantage of the scenario to bypass the formal call termination systems that fetch higher tariffs to telcos.  The calls routed through the IP networks are terminated using local SIM gateways, thus compromising the formal interconnection networks and bringing heavy losses to the telcos who have invested in building the networks. Traditionally, African countries are known to have higher interconnection tariffs compared to other regions, which further explains why such frauds are prevailing in Africa.

• The SIM Box can have SIM cards of different operators installed, so a single SIM Box can operate with several GSM gateways. The availability of SIM cards at cheaper/zero price and the lack of law enforcement over the sale of prepaid SIM cards have also favored the growth of SIM Box fraud, further. Technological advancements have also contributed to the rise in interconnection frauds. The growing sophistication around SIM Box technologies has made fraud detection difficult, which traditional methodologies are no longer able to address. For instance, there are smart devices which mimic human behavior (HBS: Human Behavior SIM Box) which makes it challenging to detect using legacy technologies.

• Traditional approaches such as CDR and TCG analysis are also proving to be ineffective in combatting this menace. The Call Data Record (CDR) analysis mainly focuses on threshold-based detection which the fraudsters evade by staying below the threshold and this leads to an increase in the high false positives associated with it. Test Call Generator (TCG) analysis, on the other hand, lags in terms of coverage and predictability aspects.

Can AI/ML be the solution to address this long-standing menace?

To conclude, the recent developments around SIM Box fraud have further aggravated the challenges faced by African telcos. With little scope for regulatory remediation and traditional approaches becoming ineffective in dealing with modern SIM Box strategies, the best way forward to prevent these attacks is by incorporating advanced technologies such as AI/ML.

This would help telcos detect SIM Box frauds quickly, as early as within 10 minutes of usage. Moreover, AI/ML can help in identifying the various hidden complex patterns which otherwise would go undetected through traditional rule-based approaches. These advanced technologies can also enable telcos to monitor the changing behavior of the SIM Box without the need for manual intervention.

These technologies represent a key weapon in the telecom regulators’ arsenal in their fight against fraud. This also becomes a vital step towards constructing sustainable and safe telecom ecosystems and would help Telcos in devising an advanced fraud detection approach to address this crisis.

To find out how you can take the next step towards fighting SIM Box fraud

Learn More

Telcos are continuously innovating and evolving their business models to counter the declining revenues from traditional services. One such business model where we see a steady increase in revenues over the last few years is Handset sales. Handset (Mobile) sales in the telecommunication industry refer to the selling of mobile phones, fixed-line devices, routers, etc. With the advent of smart gadgets, the range of devices extends to smartwatches, earphones, and headsets, and has led to the use of the broader term, Device Sales, to cover the growing range and diversity of products. One tier-1 operator in the developed market has recently reported that 22% of their revenue is from device sales in their recent financial report. While device sales contribute to telco revenues, fraudsters have quickly identified an opportunity to acquire high-value devices with almost zero upfront fees and no intention to pay in the future. This is resulting in significant fraud loss to telecom operators, and unlike traditional usage fraud, the loss in these cases can be calculated directly.

Let us look at some of the fraud methods used by fraudsters to commit device/handset fraud, especially for digital or web-based orders.

• Subscription Fraud by Identity Theft– In this method, fraudsters use a person’s real identity without their knowledge to buy goods or services with no intention to pay. Fraudsters use various methods to obtain identity information like social engineering, phishing, smishing, shoulder surfing, etc.
• Subscription Fraud by Falsifying Data – In this method, fraudsters create false/fake identity documents (synthetic ID) to purchase handsets/devices without any intention to pay.
• Account Takeover – In this method fraudsters, use the stolen details of a genuine customer using social engineering, phishing, or other techniques to take over or manipulate the customer account to order handsets/devices and
  • Get them delivered to a different address and/or
  • Change the contact number for delivery to his/her number and/or
  • Update the details to choose pickup from a storage centre using a token number sent to the updated contact number.
• Payment Fraud – In this method, fraudsters use stolen credit or debit cards or, in some instances, counterfeit cards to order or purchase handset/devices.
• Credit Muling/Proxy Fraud – This method primarily involves collusion between a genuine individual using their own identity to buy goods or services and the fraudster. The individual buys the goods or services from the operator with no intention to pay and passes the goods or services to the fraudster. In some instances, there are financial arrangements between the fraudster and the individual, and in other cases, the individual complies due to threat or forceful intimidation.

In conjunction with the above-mentioned methods, fraudsters will also try and exploit any loopholes in the existing business processes to facilitate the fraud e.g. utilizing gaps in vetting processes to avoid advanced verification checks and/or manipulating courier delivery processes to change address details or simplify the interception of the device.

In recent years, we have seen a steep increase in the number of device fraud cases with organized fraud- rings playing a significant role in these crimes. For example, in June 2019, members of a US-based fraud-gang were indicted for their part in a $19 million fraud related to stolen phones and devices. Another case from the UK saw a group of 7 people arrested for their role in a £2 Million mobile phone fraud, targeting the student community. These examples are part of a wider trend with the CFCA survey estimating that telcos are incurring fraud losses of $4.78 billion due to theft and device reselling. In addition to the fraud loss, account takeovers, and identity theft impact the telco customer experience, leading to customer churn, negatively impacting the telco brand image.

In summary, revenue from handset/device sales is increasing and is an essential contributor to operators’ overall revenue. Due to this, it is a business imperative that telcos have robust business processes and intelligent systems in place to prevent handset fraud without impacting the onboarding or upgrade processes for genuine subscribers.

Want To Know How You Can Protect Your Handset Sales Revenues With Real-Time ML Based Risk Scoring.

Schedule A Demo