Today’s digital ecosystem is witnessing rapid growth across e-commerce, fintech, health tech, edutech and other sectors. Millions of dollars are being spent on developing advanced technologies and gaining access to the market. Moreover, investors, corporations, government, and academia partners are involved in providing tech startups with the essential elements to grow and succeed in the UAE and globally. Striving for operational excellence through digital transformation in terms of RoI, time efficiency, and customer satisfaction is an essential component of any innovative framework in the digital economy.
However, this ecosystem faces the ever-prevalent threat of fraud that offsets business plans, and risk investments and essentially puts people’s lives in a tight spot. The growing complexity in the digital services ecosystem demands a future-proof approach to secure the networks and prevent revenue losses.
Fraudulent actions have spiked in telephone network functions, online banking, and e-commerce. For instance, in the UAE, a court penalized a bank and telecommunication provider concerning a SIM swap fraud, which was attributed to breaches in the security systems of both the bank and the telecommunications provider. Moreover, to combat telecom fraud from foreign destinations, one of China’s largest telephone service providers will soon restrict users in the eastern Zhejiang province from receiving international phone calls. More than half of the telecom fraud in China originates from outside the country.
Fraud Patterns
Traditional telecom fraud may include the following:
- In subscription frauds, a fraudster accesses a service with false identity details with no intention to pay the bills, all transactions are fraudulent for personal use.
- In superimposed fraud, a fraudster gains a legal account or superimposes upon the normal usage of the legal customer for cellular cloning, calling card hacking, phone misuses, etc.
- PBX hacking/fraud occurs when someone gains access to a business's PBX phone system and generates a profit from the international calls at the business's expense.
- Message previews let fraudsters access sensitive information and hack people's WhatsApp and bank accounts, among others.
Apart from the traditional cases of telecom fraud, with the emergence of multimedia applications and digital audio/video transport over IP-based networks, Session Initiation Protocol (SIP) is becoming the de facto signaling protocol standard for telecom LTE/VoLTE services. SIP is the session control protocol that can establish, modify, and terminate multimedia sessions such as Internet telephone calls (VoIP), multimedia distribution, and multimedia conferences.
Fraud detection cannot be a fix-and-forget exercise. Fraud patterns can differ drastically over time with changes in customer behaviours and traffic volumes in today’s dynamic telecom ecosystems.
Leveraging artificial intelligence
The telecommunication industry is facing slumping sales due to fierce competition and fraud-related problems. As such, telcos can benefit from leveraging AI’s powerful analytical capabilities to combat instances of fraud.
AI and machine learning algorithms can detect anomalies in real-time, effectively reducing telecom-related fraudulent activities, such as unauthorized network access and fake profiles. The system can automatically block access to the fraudster as soon as suspicious activity is detected, minimizing the damage. With industry estimates indicating that 90% of operators are targeted by scammers daily, amounting to billions in losses every year, AI application is a must for telecom companies.
Markets estimate that global AI in the telecommunication market size will reach $14.99 billion by 2027, from $11.89 billion in 2020, at a CAGR of 42.6% during 2021-2027.
Moreover, network intelligence and automation are a crucial part of the evolution of 5G, IoT and industrial digitalization. Data flow generated by the telecom industries to provide billing information, call details and network data can be routinely analysed by leveraging AI capacity. AI in real-time learns to interpret data, detects new cases, and expands its knowledge base with new behaviours that may require a reaction.
AI/ML help define the correct behaviour that will automatically be performed when specific events are detected. This approach translates into savings in implementation and maintenance. It also dramatically simplifies architecture because fraud prevention mechanisms use the same data in real-time as other processes. Above all, this process secures business operations in a completely new intelligent way.
As per the latest data released by UAE’s Telecommunications and Digital Government Regulatory Authority (TDRA), the total number of telecom subscribers in the UAE, including mobile phones, landlines and Internet stood at 21.929 million by the end of 2020. The number of mobile phone subscribers touched 16.820 million by December end from 16.707 million in November. The subscription per 100 people stood at 186.1, with mobile subscriptions accounting for 76.2% of the total subscription base.
With this growth in online activities, fraudsters are using even Google ads on Google search result pages or ads on social media channels to promote fake brands. Victims to such nefarious activities have lost money purchasing counterfeit products.
Brain power and AI
By investing in a network-level fraud mitigation system using AI, CSPs can substantially enhance fraud detection but it will require the support of the unique intelligence of the human brain. As we all know, AI can only detect and predict fraud using historical data, whereas the human mind can develop creative, logical solutions to fight fraud, without having to have much data on hand. However, the processing power of compute-heavy AI technology cannot be overlooked at any cost. What will be needed is an integrated approach to fraud management and security using AI/machine learning combined with talented and creative human minds to create the ultimate machine for CSPs to fight the menace of telecom fraud effectively.