AI has become a powerful assistant at all analytics stages. Let us cover where it helps and what critically remains with the human: 1) Where AI speeds up analytics: — UNDERSTANDING THE DATA: AI helps quickly grasp an unfamiliar set — what the columns are, what problems (gaps, strange values), what can be analyzed. — CLEANING: helps find and fix errors, duplicates, gaps, format mismatches — what usually takes most of an analysts time. — WRITING QUERIES AND CODE: AI writes SQL queries, Python code for analysis from your natural-language description ("compute average sales by month") — no need to remember all the syntax (a link to the Python-for-AI course). — ANALYSIS: helps compute metrics, find patterns, compare segments, notice trends and anomalies. — VISUALIZATION: suggests and builds suitable charts for your data and question. — EXPLANATION: formulates conclusions in clear language, helps present results for a decision. 2) What CRITICALLY remains with the human (AI is an assistant, not a replacement): — POSING THE QUESTION: what question do you ask the data and WHY (for what decision)? This determines the whole analysis. AI does not know your business context and goals — you pose the right question. — CRITICAL THINKING AND VERIFICATION: AI can err — compute wrongly, invent (hallucinations from the LLM courses), misunderstand the task. YOU must verify the results: do the numbers add up, are the conclusions reasonable, are there logic errors. You cannot trust blindly. — INTERPRETATION IN CONTEXT: what do the results mean FOR YOUR task? Correlation is not causation, a number itself is not a conclusion. Understanding the context, data limitations, what really follows from the analysis is with the human. — DATA AND CONCLUSION QUALITY: garbage in to garbage out. Assessing whether the data is reliable and the conclusions grounded is your responsibility. — DECISIONS: analytics informs decisions, but a human makes them accounting for context, risks, values. 3) The key principle: AI radically SPEEDS UP the mechanical and technical part of analytics (writing code, cleaning, computing, building charts), freeing time for THINKING — posing the right questions, critical verification, interpretation, conclusions. This is not "AI does analytics for you" but "AI + analyst do analytics faster and better, where the human directs and verifies, and AI executes". 4) Accessibility: AI lowers the entry barrier to analytics — you can analyze data without being an expert in SQL/statistics/code, describing tasks in natural language. But a BASIC understanding (what data is, what errors happen, that correlation is not causation, why to verify) is still needed — or it is easy to take a nice but wrong result for the truth. The course will give this understanding. Practical rule: use AI as a powerful assistant at EACH analytics stage (understand the data, clean, write code/queries, analyze, visualize, explain) — it radically speeds up the technical work. But THINK and VERIFY yourself: pose the right question for your decision, critically verify the AI results (errors, hallucinations), interpret in context, assess the data quality and conclusion validity. AI speeds up the steps, but analytical thinking and responsibility for the conclusions are with you. This combination — AI executes, the human directs and verifies — is modern analytics with AI.