The ASTR project aims to enhance national security and prevent telecom based cyber fraud leveraging AI and facial recognition. It consolidates subscriber data from major Indian TSPs to detect fraudulent SIMs, identity duplication, and fraud networks.

Key Objectives:
* Provide a robust facial de-duplication system to prevent multiple SIMs under fake identities.
* Identify fraudulent SIM issuances across various SIM registration methods.
* Share a list of fraudulent connections with TSPs to disconnect them.
* Share list of fraudulent connections with banks, payment wallets, and social media platforms to disengage these numbers from their respective platforms.
* Detect and analyze hotspot regions with high SIM fraud incidents.
* Assist LEAs by providing them the list of PoS which issue illegal SIM cards

Results:
The product is used intensely in last two years for PAN-India SIM data analysis and detected large number of fraudulent connections. Till date, ASTR analyzed over 1.34 billion mobile connections and detected more than 8.1 million fraudulent connections, out of which 7.7 million connections were discontinued by TSPs. 71 thousand of PoS have been blacklisted so far.

Impact:
ASTR has also been instrumental in dead body identification during large-scale disasters, criminal identification, missing persons search, and tracking absconders.

https://cdot.in/cdotweb/web/product_page.php?lang=en&catId=4&pId=67

• India

• Older persons
• The poor
• Migrants
• Remote and rural communities
• People with disabilities, women, unemployed and youth

The replicability of the solution depends on a few factors, such as:

* Technology: The underlying AI models and facial recognition algorithms are indeed replicable. With the right development resources, similar systems could be built using open-source libraries or proprietary AI solutions. Companies can implement similar solutions for SIM card verification or other identity verification systems.

* Legal and Regulatory Compliance: The solution must comply with local regulations regarding biometric data and user privacy. Different countries have different laws on data protection. Replicating C-DOT's solution would require ensuring compliance with these legal frameworks.

* Infrastructure Requirements: The solution requires strong backend infrastructure to handle image processing, data storage, and AI model training. Telecom companies would need to invest in secure data centers and possibly cloud services to deploy a similar system at scale.

* Adoption by Telecom Operators: For replication, telecom operators must be on board with adopting such solutions, and they may need to integrate them with existing systems (e.g., customer management, billing, etc.).

* AI models can be trained and optimized to handle large numbers of verifications simultaneously. This is important as telecom operators expand their subscriber base, and the solution can scale as the network grows.

In summary, while the technology itself is replicable, implementing such a solution requires careful consideration of technical, legal, and operational factors.

ASTR can be considered sustainable for a variety of reasons, such as:

1. Improved Security and Compliance

> Fraudulent SIM card activations can be reduced, preventing misuse for illegal activities.
> Automating identity verification process and meeting government regulations, reducing the need for manual checks.

2. Operational Efficiency

>By automating the verification process, human error and manual intervention can be reduced. This cuts down labor costs and speeds up verification process.
> Although the system requires significant initial investment, cost of manual verification is much higher in long run.

3. Environmental Sustainability

> The adoption of digital verification reduces paper-based identity verification, which impacts environment in terms of paper use and storage.
> Modern AI algorithms, when running on optimized cloud platforms or local data centers, can be energy-efficient compared to older legacy systems that may require extensive hardware resources.

4. Adaptability and Evolution

> AI models improve over time as they learn from more data, thus ensuring the system remains effective even as technology advances.
> Beyond SIM verification, the solution can be adapted for secure access to mobile banking services, customer support, or fraud prevention.

5. User Convenience

> The solution offers a convenient, faster way to SIM card registration, especially in regions where digital identity verification is becoming the norm.
> In countries with a high number of remote or rural populations, AI-powered verification can help telecom companies reach underserved areas where traditional face-to-face verification is difficult.

6. Data Privacy and Security

> The system can be designed to ensure that biometric data is stored and processed in compliance with local data protection laws.
> The integration of end-to-end encryption, makes the system more secure, protecting sensitive subscriber information from cyber threats.

The ASTR project aligns with the World Summit on the Information Society (WSIS) values by promoting secure digital inclusion, trust in ICTs, and the responsible use of AI in combating fraud.