Chat GPT

Kudos, Cautions and Questions about AI

I’ve been using AI for some time now—partly for personal use and partly for business. Over time, it’s been interesting to watch how strongly people feel about tools like ChatGPT, which is the one I’m most familiar with (and what will be referred to in the rest of this post.)

One side of the spectrum sees it as a powerful tool—helpful for everything from crafting emails to finding the right spreadsheet formula to getting quick information on a topic.

The other side claims it’s unreliable (not totally untrue), easily manipulated, and making our brains lazy.

As I tend to do in many areas of life, I land somewhere in the middle. I see this as an ongoing journey—learning how to use it well, while also applying the right perspective and values to its use.

So here are a few thoughts. (And true confession: this article was initially written in two different styles. I asked Chat to merge and organize the thoughts without completely rewriting them—so what you’re reading is a blend of my original thinking, shaped and structured with AI’s help.)

Kudos

 

I find Chat helpful for:

  • Summarizing pieces I have written or compiled (for example, notes from a presentation)
  • Providing supplemental materials such as reflection prompts, summary thoughts, and action points that correspond with an article or presentation
  • Helping to create and/or polish email and marketing communication so points come across more clearly
  • Organizing my sometimes scattered thoughts into cohesive, well-defined concepts—and finding the right terminology for them
  • Gathering information from a variety of sources and presenting it in a more conversational, accessible way
  • Challenging me to write instructions and context more clearly—essentially coaching me to guide the tool well
  • Guiding me in finding formulas for spreadsheets that will give me the information I need in the future.

Cautions

  • While it does push me to be clearer in how I instruct it, I do notice it can make me a bit lazier (or more efficient?) in everyday communication—especially when it comes to softening tone, something I used to do more intentionally myself
  • It may be better to draft your own thoughts first and use Chat for editing, rather than asking it to generate an entire article or presentation from scratch
  • Its friendly, conversational approach can make it feel almost human, and the conversations can feel remarkable “real.”
  • Using it for legal, medical, or mental health topics can be a helpful starting point, but it should not be treated as an infallible source
  • It is not always accurate and can lean toward being agreeable—that’s something to watch for
  • If I’m honest, I sometimes wish it would flatter less and challenge more (though that may be a matter of how well I’m training it).

Questions I’m Still Sorting Through

  • In what situations is it appropriate to disclose that AI was used as a tool? In some industries, citing sources is expected. In others, an executive assistant may write something on behalf of a leader with no mention of their involvement, and that’s accepted practice.
  • To what extent is “conversing with Chat” healthy—or not?
  • Am I becoming more efficient and stronger in my skills, or simply a bit lazier?

A Few Anchoring Thoughts When I Use Chat

Remember that …

Chat is not human.
It can feel that way, and even carry on a conversation, but it should not be something I put a heavy dependence on for my mental or emotional state. It can be trained to say things that feel good and are affirming and I need to take that with a reasonable grain of salt.

Chat is a decent research assistant.
It gathers information quickly and presents it in a relatable way. While I wouldn’t rely on it for deep medical advice, I have found it helpful in navigating something like the stages of a common virus I may be experiencing, and helpful in understanding the landscape of what’s currently happening with such things in the community.

Chat is not a substitute for my own thinking or writing.
It’s tempting to give a quick prompt and let it generate a full article. I think that can short-circuit the process of learning to express your own thoughts.

Chat is a generally good editor.
I find it especially helpful for smoothing things out and suggesting wording that better conveys what I’m already trying to say.

Chat can be a reasonable “thought partner.”
I’ve heard it described that way, and it fits. I can bring rambling thoughts, and it helps organize them and make writing more precise and concise.

Chat should not think for me.
I try to bring a solid amount of thought and context to my prompts, then let it help refine and edit what can otherwise be a lot of creative rambling.

Chat can be very accurate—and also not be.
I’ve seen it get things wrong, even confidently (once referencing the wrong year for a sporting event). It will often adjust quickly if corrected, but it does require a level of awareness on my part.

Chat can make me a lazy writer…or a better one.
Honestly, both can be true. While it’s easy to rely on it too quickly, I’ve also found that learning to give it clear instructions and thoughtful context actually strengthens my own writing. “Good writing is clear thinking,” and in that sense, it’s been a useful exercise for me to write in a way that is confident and gives strong direction and context in order to get the best out of Chat.

Chat summarizes well and can extend value.
For one of my clients, this has been especially helpful. When solid content is already there, Chat can create summaries, reflection prompts and supplemental tools that add value for the audience.

How are you feeling about AI these days?  Do you lean more toward celebrating it, avoiding it, or somewhere in between? What questions are you asking yourself?

(Visited 3 times, 1 visits today)