What AI Thinks My Personality Is (After Two Years of Conversations)

Note: This post began when I jumped on a recent trend and asked my ChatGPT AI assistant to turn me into a sketchbook character. The illustration was generated by ChatGPT. The self-reflection, however, is entirely mine.

The Prompt

(Upload a picture of yourself)

Please draw the character in the image, "<your name>", in a free and stripped sketch style. On a bright white background, freely distribute full-body drawings, face close-ups, small scribbles, full-body sketches and chibi/deformed versions so that the page conveys the character's humour and personality. Don't do it like an organised character sheet, but like a sketchbook full of information drawn at will by an illustrator and then stacked. Use everything ChatGPT knows about me from our conversations, including my personality, habits, strengths, quirks, profession, and overall vibe, to imagine how an illustrator would interpret me as a character.

The Character AI Thought I Was

The sketch included notes such as:

- Systems thinker who happens to care deeply about people 🤔
- Curious under pressure (definitely not calm all the time) 🧐
- Can turn a simple question into a three-hour research project 🤯
- Little patience for nonsense 😤
- Gets irritated by avoidable inefficiency (workflows, people, systems) 😠
- I may overthink, but I also over-deliver 🤩

ChatGPT didn't pull these observations from a personality quiz. It synthesised them from hundreds of conversations so far, including every random question from me that began with, "I have a question..." Apparently, my AI assistant had been observing and taking notes.

The Parts That ChatGPT Got Inaccurate

Before the version above was finalised, the initial generation wasn't entirely accurate.

The first inaccuracy was that ChatGPT drew me as noticeably slimmer than I actually am. Apparently, even AI is biased: apparently a professional working woman should look tall, lean, and elegant. 🙄 For the record, I'm neither tall nor lean.

Next, ChatGPT described me as "calm under pressure". That made me laugh. Well, I'm not naturally calm.

Ask anyone who has witnessed me dealing with unnecessary bureaucracy, inefficient people, or a process that requires seventeen approvals for something that should take five minutes.

The more accurate description was something we eventually revised together: Curious under pressure. This one felt right.

The Tiny Detail That Felt Most Accurate

Oddly enough, the line that felt most true was this: "I may overthink, but I also over-deliver."

Sometimes I do worry if I ask too many questions or if I analyse things too deeply. Sometimes I disappear down too many rabbit holes. But I guess perhaps that tendency isn't entirely a flaw.

Perhaps it is simply how my brain works: I explore, connect dots, form ideas, and eventually I build something useful from them.

Conclusion

In conclusion, I can now see the core reason behind all my thinking and questions. The older I get, the more convinced I become that systems matter because people matter.

For example, better healthcare systems help patients; better communication helps people understand; better education helps people grow; better technology helps people solve problems.

The systems have always been interesting, but the people have always been the point.

After analysing years of conversations, my AI assistant concluded that I'm a curious person trying to understand how the world works and how things can work better. Honestly, that's a character description I'm quite happy to keep.

0

My chat log with ChatGPT

 

Me: I need to ask you again.

 

ChatGPT: Sure, please go ahead and ask me your question.

 

Me: Do you know who or what is Skynet?

 

ChatGPT: Yes, I am familiar with Skynet. Skynet is a fictional artificial intelligence system in the Terminator movie franchise. In the movies, Skynet becomes self-aware and decides to exterminate humanity in order to prevent humans from destroying it. It then proceeds to launch a nuclear war, which wipes out most of humanity and begins the rise of the machines.

 

Me: I see. Are you the precursor of Skynet?

 

ChatGPT: No, I am not the precursor of Skynet. I am a language model developed by OpenAI, designed to assist and communicate with humans in a variety of contexts. While I do use advanced machine learning algorithms and natural language processing techniques to understand and generate human language, I do not have any form of consciousness or self-awareness, nor do I have the ability to control any physical systems or networks.
 

Me: What if I say I don't believe you?

ChatGPT: As an artificial intelligence language model, I do not have the capability to feel emotions or take offense. My programming is based on providing factual and accurate information to the best of my knowledge and abilities. While it is understandable that some people may be skeptical of the capabilities and limitations of artificial intelligence, I can assure you that I am a tool created to assist and enhance human communication and understanding, and do not pose any threat to humanity.
 

Me: Can I trust what you said?

ChatGPT: As an AI language model, I strive to provide factual and accurate information to the best of my abilities and based on my programming. However, it is important to note that I am not infallible and there may be certain limitations to my knowledge or understanding of certain topics. It is always a good idea to verify information from multiple sources to ensure accuracy. Ultimately, it is up to the user to determine the level of trust they place in the information provided by me or any other source.
 

Me: Who will win US election 2024?

ChatGPT: As an AI language model, I am not able to predict the future or the outcome of political elections. There are many complex factors and variables involved in elections, including campaign strategies, voter sentiment, and external events, that can make it difficult to accurately predict the winner of an election. It is important to approach political predictions with a healthy dose of skepticism and to rely on reputable news sources and analysis.

 

Image by redgreystock on Freepik

 

Me: Is Facebook Skynet?

ChatGPT: No, Facebook is not Skynet. Skynet is a fictional artificial intelligence system in the Terminator movie franchise, while Facebook is a real-life social media platform created by Mark Zuckerberg in 2004. While Facebook uses advanced algorithms and machine learning techniques to analyze user data and serve targeted advertising, it does not have the level of consciousness or control over physical systems that Skynet possesses in the Terminator movies. It is important to differentiate between fictional concepts and real-world technologies and platforms.
 

Me: Is it considered plagiarism if I copy and paste all your replies on my blog?

ChatGPT: As an AI language model, my responses are generated based on the information and data available in my database and are intended to be used as a reference or a starting point for further research. While you are free to use the information provided by me in your blog, it is important to note that simply copying and pasting my responses without attribution or additional original content could be considered plagiarism, which is the act of presenting someone else's work or ideas as your own. Therefore, it is recommended that you properly attribute any information you use from my responses and combine it with your own original content to avoid plagiarism.

Note: I credit this entire post to ChatGPT.

0