Chapter 1: Understanding ChatGPT's Impact

This discussion isn't centered on whether ChatGPT will make you wealthy or lead to global chaos. Instead, it focuses on practical examples of how AI technologies can significantly enhance business operations. ChatGPT, an AI chatbot introduced by OpenAI in November 2022, is powered by the GPT-3.5 series of advanced language models, refined through supervised and reinforcement learning methods. — Wikipedia [1]

My prior knowledge of ChatGPT was limited, but a recent video by renowned psychologist Jordan Peterson piqued my interest. He predicts that ChatGPT will revolutionize industries, potentially leading to significant job losses. While it’s true that AI may replace certain roles, the notion of a complete upheaval might be overstated.

Additionally, it's crucial to recognize that AI can perform many tasks that humans cannot. To illustrate this, I examined a data analysis scenario using BigQuery SQL. Could ChatGPT actually assist me in generating useful scripts? It's plausible that data analysts or business professionals could enhance their analytical capabilities with AI support. Given the many unfilled positions in this field, individuals seeking to perform data analysis might find immediate assistance through ChatGPT rather than waiting for responses on forums like Stack Overflow or Reddit. I opted for BigQuery SQL because it's the tool I use most frequently; however, this technology can also be applied to various programming languages and frameworks.

Use Case: Data Analysis with BigQuery SQL

Imagine you’re a business user aiming to analyze data with BigQuery. As organizations increasingly adopt data-driven strategies, employees from diverse backgrounds are utilizing (self-service) BI tools or sophisticated platforms like Google BigQuery. Often, you may know the question you want to ask but struggle to formulate the right code. Let's see if ChatGPT can provide assistance.

First Inquiry:

How can I pivot in BigQuery SQL? Can you create a sample script?

Here’s the response from ChatGPT:

WITH data AS (

SELECT 'A' AS category, 1 AS value UNION ALL

SELECT 'A', 2 UNION ALL

SELECT 'A', 3 UNION ALL

SELECT 'B', 4 UNION ALL

SELECT 'B', 5 UNION ALL

SELECT 'C', 6

)

SELECT * FROM data

PIVOT (

SUM(value)

FOR category IN ('A' AS A, 'B' AS B, 'C' AS C)

)

The result is impressive; it executes flawlessly and even provides a comprehensive explanation of the script and the PIVOT operator. The code is highlighted appropriately, relieving me from the task of formatting it myself. This script generates a sample dataset with three categories (A, B, and C) alongside their respective values. The PIVOT operator aggregates the values for each category, producing a table with three columns—one for each category—showing their summed values.

To utilize the PIVOT operator, you'll need to define:

• The input table (in this case, 'data').

Second Inquiry:

Could you create a BigQuery SQL script that generates random numbers?

ChatGPT provided the following:

CREATE TEMPORARY TABLE random_numbers AS

SELECT CAST(RAND() * (100 - 1 + 1) + 1 AS INT64) AS random_number

FROM UNNEST(GENERATE_ARRAY(1, 10))

WHERE RAND() <= 1;

This script allows you to generate 10 random numbers ranging from 1 to 100. You need to specify:

• min: Minimum value of the random number range.
• max: Maximum value of the random number range.
• num: Total random numbers to generate.
• probability: Likelihood of generating a random number (useful for generating a subset).

Conclusion

As demonstrated, ChatGPT effectively generates functional code, and this capability extends to various programming languages and systems. The objective of this article isn't solely to showcase ChatGPT's abilities; it's to highlight its implications for professionals. This AI can facilitate script creation, particularly in data evaluation. In an era where skilled professionals are in short supply, tools like ChatGPT can be invaluable for those looking to broaden their expertise. They enable individuals to analyze data without requiring extensive programming knowledge. Moreover, this technology is versatile—employees beyond the Data Science domain, such as those in Marketing, can leverage ChatGPT for tasks like generating CSS scripts for website modifications.

Ultimately, such AI systems are poised to simplify workflows in the future, offering solutions to challenges and serving as excellent resources for learning programming languages.

Sources and Further Readings

[1] Wikipedia, ChatGPT (2022)