"Securing AI: Navigating Privacy, Security, and Compliance"

This topic explores the critical intersection of privacy, security, and compliance in AI development. It highlights the challenges organizations face in protecting sensitive data while ensuring regulatory adherence, and discusses strategies for implementing robust security measures to safeguard AI systems.

Data Science professional with 17 years of experience. Works on Data Science/Machine Learning using Python/R.

"Human and AI collaboration: Governance Strategies for Trustworthy AI and Safe AI system’s"

Governance strategies are essential for ensuring trustworthy and safe AI. These include regulatory frameworks, ethical design, and ongoing monitoring to promote transparency and accountability. Collaboration between policymakers, technologists, and stakeholders is key to building public trust and aligning AI with human values.

I am an AI Engineering Manager at IBM with a PhD in AI and Data Analytics and over 17 years of experience.

I am responsible for delivering innovative solutions for Apple accounts and managing large-scale teams and projects. My expertise includes text and numeric data solutions, using techniques like NLP, generative AI, and geo-spatial analysis. With 11 publications, I am also a guest speaker and passionate about solving real-world problems through AI and data-driven methods.

"From Smart Automation to Adaptive Intelligence: Generative AI Agents Transforming Business"

In today’s dynamic business landscape, organizations face increasingly complex decisions that demand agility, intelligence, and strategic foresight. Generative AI agents offer a powerful solution, evolving from simple automation tools to adaptive systems capable of dynamic, multi-layered problem-solving. This presentation explores two game-changing approaches in AI: Strategic AI Agents, ideal for complex, modular decision-making environments, and Adaptive Agents, which adjust their responses in real time based on task complexity, customer interactions, and feedback. Through frameworks like Agent RAG, Fusion RAG, and Graph RAG, and cutting-edge advancements in adaptive learning, we’ll examine how these agents empower businesses to achieve real-time personalization, optimize resources, and drive smarter, decentralized decisions. Join us to discover how strategic and adaptive generative AI agents can revolutionize your business, enabling it to navigate complexity, enhance customer engagement, and stay competitive in an AI-driven future.

Ph.D. Research Scholar(NeuroSciences) in Data Science by Education, a Data Scientist by Passion and Profession, and a Leader by experience. Blended 15 years of work experience in Technical, Functional, and Leadership skills in the areas of Data Science, Machine Learning, and Deep Learning with a strong Research mindset, which helped me set up frameworks and solutions from scratch. An Influencer, Motivator, and Supporter of all the Good Causes.

"Building scalable AI systems"

This talk addresses the challenges of building large-scale AI systems, especially with Large Language Models (LLMs) and generative AI, divided into two parts:

### Part 1: General Challenges in Scaling AI

1. **Computational Resources**: Significant computational power is required, driving up costs.
2. **Data Quality**: Acquiring high-quality, domain-specific data is complex and resource-intensive.
3. **Fine-Tuning**: Adapting LLMs for specific domains demands extensive time and expertise.
4. **Explainability**: Transparency becomes crucial as models increase in complexity.
5. **Integration**: Merging AI models with legacy systems poses technical challenges.
6. **Model Updates**: Regular retraining is costly and labor-intensive.
7. **Regulatory Compliance**: Meeting industry regulations is essential.
8. **Talent Scarcity**: A shortage of skilled professionals hampers scaling efforts.

### Part 2: Challenges in the Legal Domain

Applying LLMs in the legal field introduces unique challenges:
1. **Data Volume**: Vast amounts of complex data complicate analysis.
2. **Data Quality**: Legal documents must be cleaned and standardized.
3. **Handling Precedents**: Legal AI needs to accurately interpret past rulings.
4. **Ambiguity**: The nuanced nature of legal language challenges LLMs.
5. **Jurisdictional Differences**: Models must adapt to varying laws.
6. **Adversarial Scenarios**: Balancing opposing legal arguments is complex.
7. **Continuous Updates**: Legal AI must stay current with new regulations.

### Startup Strategies

To overcome these challenges, the startup adopted strategies such as:
1. **Focusing on a Niche**: Concentrating on a specific area of law for manageable development.
2. **Leveraging Open-Source Models**: Using pre-trained models to reduce costs.
3. **Efficient Data Pipeline**: Streamlining data collection and cleaning.
4. **Collaboration with Experts**: Partnering with legal professionals for accuracy.
5. **Public Legal Databases**: Utilizing publicly available resources for training.
6. **Robust Privacy Measures**: Implementing strong protections for compliance.

As the AI field evolves, ongoing adaptation to new challenges is crucial.

Prof S Chandrasekhar is Senior Prof & Director Business Analytic at IFIM Business School B “Lore since Nov 2013.

He was chair Professor (July 1998-Feb 2013) & officiating Director at FORE School of Management, New Delhi (April 2009 to Jan 2010). He joined FORE School in 1988 as Senior Professor. He also headed the software development at Fore School of Management which are developing proto type of products in the area of Business Intelligence, Risk Management, and Customer Relations Management. He has a total of about 34years of experience in R&D, Academic & Industry in the area of Quantitative Techniques & IT. Taught courses in the area of IT, QT, Adv. Market Research at Post Graduate level.

Prior to this he worked at Indian Institute of Management, Lucknow for about ten years(1988-98) as Professor in the area of Computers & information Systems. Member Secretary of IIM L Governing Board for about three years.

"Advancement in AI Technologies. From GPT to general AI"

A seasoned Senior Generative AI Engineer with extensive experience leading transformative AI projects at prestigious organizations like Ascendion and Jio Platforms. Expertise in the end-to-end development and deployment of AI/ML solutions, including large language models (LLMs) and privacy-centric AI frameworks. Recognized for designing and implementing cutting-edge AI solutions that enhance operational efficiency and customer experience. Proficient in Agile methodologies, with a proven track record of steering cross-functional teams towards achieving technological milestones. Published author on AI model optimization and active participant in AI industry workshops and conferences. Holds advanced certifications in Big Data, Python, Machine Learning, and AI from leading institutions.

"Agentic AI: Building Systems That Reason, Adapt, and Act"

Neeraj Gudipati is a Scientist at Conduent Labs (formerly Xerox Research Centre). He is a multifaceted professional deeply invested in advancing urban planning and transportation solutions through innovative research and technology integration. Within Conduent Labs, he focuses on research in urban representations, multi-modal mobility planning, and urban space management to foster sustainable cities. His work as an individual contributor intersects with advancing sustainable, healthy, and equitable cities through collaborative knowledge co-production and advanced modelling techniques in urban infrastructure, transit, transportation, and policy. Noteworthy projects in this domain include the development of a patent-pending system leveraging augmented reality and multi-source sensor integration for constructing digital twins of curb spaces. He has led the creation of a patented system enabling immersive navigation for enforcement routing using geo-spatial augmented reality and cartographic systems, aimed at capturing violations and intuitively visualizing curb regulations in augmented reality. Neeraj’s contributions at Conduent Labs have resulted in the publication of 3 patents and the filing of 6 patents, underscoring his commitment to innovation and impactful research outcomes.

He received his Bachelor's degree from the Indian Institute of Information Technology and was previously a pre-doctoral researcher at Harvard and MIT Media Labs. During his time at MIT, he founded a think tank named MITTI, an MIT TATA Centre Initiative to help smallholder farmers with sustainable agriculture using low-cost sensors and Soil-AI. His work revolves around three core areas: Information Architecture & Data Reliability, Deployable Machine Learning, and Decision Systems and Contextual Dynamics. Neeraj's efforts are dedicated to addressing challenges such as data biases, resource constraints, and local context understanding, with a focus on delivering equitable and impactful solutions.

"Scope of gen AI in digital supply chain"

Over 12+ years of expertise in Supply chain Management and Product Life cycle Management of Automative, Heavy Engineering, Earth Moving equipment, Electronics and computers industries.