In today’s hyper-connected digital world, data security has become one of the most urgent global challenges. From hospital imaging systems to financial platforms and government infrastructures, billions of images are stored and transferred daily—making them vulnerable to cyberattacks, unauthorised access, and digital manipulation. To address these challenges, an important session of the STEAM Conference 2025 was held at Ali Auditorium, under the Computer Science (3), Physics, and Mathematics track, focusing on a groundbreaking approach: image encryption inspired by chaotic fluid flow dynamics.

The session was chaired by Dr Muhammad Arif (GM RND, Chenab Engineering, Faisalabad), who emphasised how advanced mathematical models, particularly chaos theory, can reshape the future of cybersecurity. Supporting him as Co-Chair, Dr Uzair Saeed from The University of Faisalabad highlighted that combining physics-driven systems with modern cryptography has the potential to produce encryption frameworks that are not only secure but also computationally efficient for real-time digital applications. Together, they set the stage for a thought-provoking discussion on the evolving landscape of secure image communication.

Keynote Insight: Ethical and Efficient Use of Generative AI – Ahmad Zeeshan (IIUI, Islamabad)

The keynote talk was delivered by Ahmad Zeeshan, who presented insights on the “Efficient and Ethical Use of Generative AI.” He explained that generative models are becoming central to creative industries, smart automation, and scientific research—yet their use raises serious questions about misinformation, surveillance, and data exploitation. He stressed that any encryption model of the future must integrate ethical safeguards to ensure that AI systems do not inadvertently expose sensitive visual data. His talk created a strong connection between generative AI and the need for secure image transmission frameworks.

Invited Talks Strengthening the Cybersecurity Vision

AI-Powered Formative Assessment for E-Learning

Saman Iftikhar, Arab Open University, Saudi Arabia

Saman presented FASTER-OS, a formative assessment system powered by AI for operating systems courses. She demonstrated how educational platforms increasingly rely on digital data, requiring robust encryption to protect student identities and learning analytics.

Laser-Induced Breakdown Spectroscopy for Disease Detection

Shakeel Saeed, The University of Faisalabad

Shakeel introduced an advanced diagnostic method using LIBS integrated with machine learning to detect female diseases. He highlighted that medical imagery is among the most sensitive digital assets—making secure image encryption a critical component of modern healthcare technologies.

Blockchain for Protecting Generative AI Prompt Data

Bushra Sana Idrees, The University of Faisalabad

Bushra showcased a blockchain-enabled secure access system designed to protect generative AI prompt data—a new but rapidly growing concern. Her research demonstrated how decentralised ledger systems ensure tamper-proof storage, complementing encryption models that defend against data theft.

Confidential Computing in Serverless Clouds

Tanzeel Zaidi, The University of Faisalabad

Tanzeel explained the future of multi-party machine learning, where different organisations jointly train AI systems while keeping their data confidential. He emphasised that secure encryption, such as the chaotic fluid flow approach, is essential for enabling safe and collaborative learning environments.

NEURO-ARIEL: Emotion-Aware AR and AI Framework

Rabia Kanwal, The University of Faisalabad

Rabia introduced NEURO-ARIEL, a cutting-edge AR and LLM-assisted learning system designed for autism-inclusive classrooms. Capturing emotional cues through AR devices raises privacy challenges, making secure image encryption vital to ensuring data protection for vulnerable learners.

Why Chaotic Fluid Flow Dynamics Matter in Encryption

Much like turbulent currents in fluid mechanics, chaotic models exhibit randomness, unpredictability, and extreme sensitivity to initial conditions. These properties make them ideal for designing encryption schemes that are tremendously difficult to reverse-engineer.

The presenters discussed how this technique can:

  • Strengthen defence against brute-force and statistical attacks
  • Enable real-time encryption for communication networks
  • Improve security for biometric and medical imaging systems
  • Support confidential computing and federated AI workflows

By merging physics, mathematics, and computer science, the proposed framework opens new pathways for building ultra-secure encryption models that can withstand modern cybersecurity threats.

Conclusion

The session delivered one clear message: as digital ecosystems grow more complex, security must evolve faster than the threats themselves. With leadership from Dr Muhammad Arif and Dr Uzair Saeed, and enriched by diverse contributions from Ahmad Zeeshan, Saman Iftikhar, Shakeel Saeed, Bushra Sana Idrees, Tanzeel Zaidi, TUF, and Rabia Kanwal, the discussion highlighted a revolutionary direction—using chaotic fluid flow dynamics to design next-generation image encryption systems.

By integrating ethical AI principles, blockchain security, confidential computing, and domain-specific innovations, the future of digital protection is moving toward systems that are secure, intelligent, and scientifically inspired.