AI and Data Science: Leveraging Responsible AI, Data and Statistics for Practical Impact
A collaboration between Astana IT University and MIT IDSS
Fundamentals of AI for Leaders in Public and Private Sectors
Professor Munther Dahleh, Founder and First Director of MIT IDSS, will personally conduct a 3-day intensive course at Astana IT University on June 17–19 as part of AI Week.
3-day bootcamp | Start date: June 17, 2026
Language of instruction: English
About the Program
This three-day course provides a comprehensive view of AI, examining its benefits and challenges as it is deployed across various sectors of society. It presents a unified perspective on the full AI lifecycle, exploring how different components interact to create viable solutions for societal challenges. Additionally, the course addresses the regulatory landscape, focusing on strategies to mitigate unintended consequences such as system failures, biases, discrimination, and ethical concerns while maximizing AI’s overall societal benefits. Delivered in a discussion-based format, the course allows ample time for participants to develop and refine their own ideas. An effective AI strategy consists of four key components: infrastructure investment and maintenance, skilled workforce development, research and entrepreneurship, and governance and regulation. Implementing such a strategy requires strong collaboration among stakeholders, particularly business leaders, government officials, and educators. While mapping an AI strategy—whether for a government or a business—is a complex undertaking, this course simplifies the process by providing a structured framework to analyze AI technologies.
Who Is the Program For? (Executive Audience)
- C-level executives — CEO, CTO, CFO, CDO
- Civil servants and policymakers
- Strategy and Innovation Leaders
- Team leaders in regulated industries
- Heads of Digital Transformation
- Those who make decisions on AI implementation
What Makes the Program Different (Executive-Level Framework)
- Unified Framework
Технологии политика и стоатегия в одной целостной рамке. Лидеры получают объединённый взгляд на AI-внедрение, а не разрозненные инструменты. - Decision-First
Approach
Focus on decision-making, not technical depth. The course prepares leaders to make AI-driven decisions, not to write code. - Real-World
Risks
Bias, misinformation, system failures, and regulation. The program teaches participants to identify and reduce key AI risks in practice. - High-Stakes Environments
Designed for leaders working in complex, high-stakes environments where the cost of error is significant: top management, government bodies, and regulated industries.
Program Curriculum
Module 1
• What is AI; the full lifecycle (Example: Health Care)
• High level: What is AI, ML, LLMs, Gen AI, agentic AI
• Brief History of AI (1950—2025): The 4 revolutions of computing and AI
Module 2
• What is Data Science (methodology, Data Quality, Problem formulation, causality)
• ML overview: Regression, Decision Trees, Random Forests, Deep NNT, reinforcement learning
Module 3
• Matrix completion:
Example: Recommendation Engines, policy assessment
• Random Models
Example: Time series in engineering, finance, Pandemics modeling, Gene Folding
• Transformers and Large Language Models
Example: Large Language models (LLM) and prompt engineering in business applications
Module 4
• Agentic AI: How it works
Example: Demonstration of chatbots and explanation of how they are designed
Module 5
• Issues of Bias, Causality, and ethical solutions
• Data Governance:
- Digital Platforms, Electronic Health Records
• GDPR: Data regulation, what framework is adopted for property rights?
Example of unintended consequences (innovation, failure in some scenarios)
• Legal aspects: Externalities, Responsibility for failure
Example: discrimination due to data bias (explain source of sampling bias)
Example: Externality— Legal aspect (Golden State Murderer)
Module 6
• Back to the AI life cycle, interdependencies between components
• Data/Algorithms dependency
Example of bias and discrimination in Digital Platforms
Antitrust in e-commerce platforms
• Algorithm/people dependency
Example of worse outcomes with ML assistance to doctors
Example: Mono-culture such as CrowdStrike or AI Assessment tools
• Ethical vs Legal
Example: What is considered private? Data security
Example: Algorithms cannot make subjective decisions (self driving cars)
About Professor Munther A. Dahleh
William A. Coolidge Professor, MIT EECS
Founder and First Director of MIT IDSS
Member of the Laboratory for Information and Decision Systems (LIDS)
Internationally recognized researcher in networked systems, control theory, decision-making, and systemic risk. He led MIT IDSS from its founding in 2015 until 2023.
Research interests: networked systems, social networks and opinion dynamics, systemic risk, transportation systems and infrastructure resilience.