My Month in Review with Chat-GPT

Chat GPT, the remarkable creation of OpenAI, possesses a wide range of capabilities that both impress and amuse. Let's take a closer look at what this text-based conversational AI excels at and where it may fall short. For the past month, I've used Chat-GPT within non-work related, hypothetical situations to assess its much lauded abilities from a data/ solution engineering standpoint.

Creating Sample Information: Unleashing Creativity

When it comes to generating sample information, Chat GPT is a powerhouse of creativity. It can spin captivating stories and provide imaginative scenarios that leave you entertained and engrossed. Whether you need a fictional tale about talking animals or an intriguing anecdote, Chat GPT will weave an enchanting narrative that tickles your imagination. In one case, I used chat-gpt to automatically, and thematically generate a color name for 30 generated colors, which all aligned to specific criteria and musical based compatibility. It not only created a beautiful, deep color palette but named the colors in a creative way which saved me a few hours of tinkering. By having the ability to generate a data table with color names and hex codes, my hypothetical low-fi prototype would have been off to a strong start....now if only it could plug this all into Figma...

Despite it's success it became essential to remember that Chat GPT's output is based on the patterns it has learned from vast amounts of text. This means that while it can conjure up fascinating narratives, it may occasionally provide information that is not accurate or reliable. It's always a good idea to fact-check any information generated by Chat GPT to ensure its validity. Not being able to display the hex codes actual colors was an immediate limitation.

Providing Code Snippets: Assisting Developers

Chat GPT proves to be a helpful companion when it comes to providing code snippets. It can offer suggestions and solutions to programming challenges, making it a valuable resource for developers in need of a fresh perspective. Whether you're troubleshooting an issue or seeking an alternative approach, Chat GPT can provide insights that spark ideas and propel your coding endeavors forward. For this use case, I used a wide array of languages including R, Python, DAX, SQL and more. I was shocked at how clean the code was for simple queries. It was able to produce code (which can be easily copied into VS) which include helpful stepwise notation commented out appropriately for each language.

However, it became immediately important to exercise caution when relying solely on Chat GPT's code suggestions. While it can offer valuable guidance, it doesn't have the same level of understanding as a human programmer. It made mistakes....lots of them. What I found useful is building upon the code SLOWLY. Starting with a description of what I was trying to achieve, giving additional information and adding functionality in a stepwise motion. This required me to have a very strong understanding of how a real developer would tackle the issue. For right now, Chat GPT is no substitute for a human programmer, however, in the future, it's very possible they get closer. My overall sense is, it's wise to review and validate the code snippets provided by Chat GPT before implementing them in your projects. Remember, human expertise is still crucial in ensuring the reliability and functionality of your code.

Providing Data Recommendations: An Analytical Ally

When it comes to data recommendations, Chat GPT can be an invaluable ally. It can analyze patterns, trends, and correlations within datasets, providing insightful recommendations and interpretations. Whether you seek information on market trends, climate patterns, or any other data-driven inquiry, Chat GPT can help you explore and make sense of the information at hand. For this example I provided some sample data to Chat GPT and had it walk me through recommendations for statistical analysis. What was GREAT is that it could walk me stepwise through SPSS procedures and provided insight into how I may interpret the statistical results. Overall, this is a REALLY complex task and while Chat GPT didn't do the actual analysis of the data due to limitations within input, it help provide ideas on where I may go next, and which buttons to click.

I found this to be a great helper from an instructional stance, but once again, not a replacement of having some idea of the "what" that needs to be done.

Final thought: Chat GPT is a captivating AI companion that excels in creating sample information, providing code snippets, and offering data recommendations. Its creative abilities, programming assistance, and analytical insights can be immensely valuable. However, it's crucial to remain aware of the limitations inherent in relying solely on AI-generated content. By using Chat GPT as a tool, complemented by human expertise and verification, we can harness its strengths and navigate its limitations effectively. I think there will always be a person behind the scenes who can organize the work, understand the "why" and provide the methodical context-based insight into providing the instruction.

Verdict: Yes, No, Maybe

Did it help me save time? Yes....no...I'm not sure.

In some cases, yes. In fact- it even wrote some of this article. However, buyer beware. In other instances, such as coding and data, it became a huge time suck. I probably spent more time troubleshooting and debugging someone else's work where I would have otherwise saved time just doing it myself, the old fashioned way.

The "WHEN" you should use AI will become the next great technological critical thinking skill that we will all need to begin considering/ preparing our staff for.