Explained | Will AI tools help detect telecom fraud?
2023-05-27
The Hindu
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
The Department of Telecommunications used AI-based facial recognition (ASTR) to identify and deactivate over 17,000 SIM cards in Bihar and Jharkhand that exceeded legal limits or were linked to forged documents, aiming to curb telecom fraud. The move raises concerns about privacy and lack of data protection regulations.[AI generated]
Why's our monitor labelling this an incident or hazard?
The event involves the use of an AI system (facial recognition technology) in an operational context to detect and prevent telecom fraud, which is a form of financial and cyber harm. The AI system's use has directly led to the identification and blocking of fraudulent SIM cards, thus mitigating harm. However, the article also discusses risks related to privacy violations, potential misidentification (which could lead to wrongful harm), and the absence of data protection laws, which are violations or risks to human rights. Since the AI system is actively used and has caused direct impact (both positive in fraud prevention and potentially negative in privacy and misidentification risks), this qualifies as an AI Incident. The concerns about privacy and consent, while significant, do not negate the fact that harm related to rights and privacy is occurring or is very likely occurring due to the AI system's deployment without adequate safeguards.[AI generated]