To succeed, foster a culture of awareness and resilience!

Dr. Arindam Sarkar 
Head of Department of C S & Electronics, Ramakrishna Mission Vidyamandira- Howrah  

While we handle vast volumes of data, the real challenge lies in safeguarding and managing it effectively. He shared that cybercriminals exploit weaknesses in systems, manipulating sensitive information like land records and tampering with biometric authentication methods. To counter this, we’ve begun shifting towards Virtual IDs (VIDs), believing them to be more secure. However, this confidence is undermined by the risks posed by malicious apps we often download for convenience, which can compromise our data. Generative AI, supported by large language models (LLMs), is a double-edged sword in the realm of cybersecurity. It operates on deep learning principles, primarily using two components: a generator and a discriminator. The generator creates synthetic data, while the discriminator evaluates its authenticity. This mechanism is particularly valuable in fraud detection and analysis, bridging gaps between reported incidents and actual crimes. It allows organizations to uncover hidden patterns and enhance their security frameworks.

The next leap in AI-driven security is federated learning (FL), which replaces traditional centralized machine learning models. FL enables multiple organizations to collaboratively train machine learning models without sharing sensitive data, thus preserving privacy and strengthening security. This is a game-changer for industries hesitant to move their data into cloud environments due to confidentiality concerns. By prioritizing FL, organizations can harness the power of AI while safeguarding their most valuable asset data.