Broker Check
The Wooden Nickel:  Pickle Sharks

The Wooden Nickel: Pickle Sharks

April 17, 2024

“In a quaint seaside town, a peculiar legend whispered of a pickle-loving shark that prowled the depths. Local kids dared each other to swim near the old pier, where Stephen King once sat, penning tales of horror and suspense, weaving the shark's appetite into his chilling narratives. Years later, when King revisited the town, he found inspiration anew, writing a spine-tingling story where the pickle-loving shark became the unsuspecting villain in his latest bestseller.”

What the heck was that, you ask?  It was a story written by a robot.

I asked Chat GPT (the Artificial Intelligence Engine, AI) to “make up a three-sentence story about a shark, a pickle, and Stephen King”.  I must admit that the story made me laugh out loud.  On the surface, the story seems useless, but maybe there’s utility in AI. 

I tried again, this time with something more relevant:   “Hey Chat GPT, give me ten bullet points on running an efficient financial planning practice.”  The AI gave me ten very well-thought-out bullet points.  They were specific, relevant, and useful.  The answer I got from AI would be a decent kick-off to a future blog post.

I can see how AI could be used by a person to get her creative mind working, pulling up all kinds of relevant subjects to be woven into an article or a speech.

 

Humor, it seems, is tougher.  “Hey Chat GPT, tell me a joke about an alarm clock, the New York Yankees, and whipped cream”.  You be the judge of how funny it is: 

“Why did the New York Yankees hire an alarm clock to manage their team?

Because they heard it was great at "whipping" players into shape, but they didn't expect it to hit snooze as much as they do! Maybe they should replace it with whipped cream—it's smoother and less likely to drop the ball!”

Ugh.  That is a horrible joke.

After trying to create some more jokes with AI, I conclude that it LOVES puns. A play on words requires a large vocabulary and knowledge of the language.  A pun is kind of formulaic.  That’s probably why the AI uses them, but they aren’t very funny.

 

I then asked to make a haiku about Taylor Swift:

“Swift's melodies flow,

Lyrics dance with heart's echo,

Taylor's tales aglow.”

Hmm, not sure this is going to win any prizes, but it’s certainly a haiku.

The more use the Chat GPT AI, the more I feel like I am the one who is becoming more creative.  It’s kind of hard to come up with interesting things to ask it to do, but the exercise has gotten my creative mind firing.  After all, in what possible circumstance would I need a haiku about Taylor Swift? 

Actually, I posed that question to Chat GPT.  “Hey Chat GPT, when would I need to have a haiku about Taylor Swift?”

Chat GPT:  “You might need a haiku about Taylor Swift if you're creating a themed poetry anthology, planning a tribute event, or simply want to express admiration for her music and influence in a concise and poetic form.”  Well ok then!

 

Chat GPT makes mistakes. For instance, I asked it to list the astronauts who landed on the moon in order of height.  It attempted to do so, but it had Neil Armstrong (180 cm) out of order.  I told Chat GPT it made a mistake (without saying what the mistake was), and it apologized and fixed the error. 

Then I asked it to list the same people by age, youngest to oldest, when they landed on the moon.  The AI couldn’t figure out how to do this correctly, although it certainly acted like it had given me the right answer.  Chat GPT is certainly confident in its answers!  After five unsuccessful tries, Chat GPT essentially gave up while apologizing profusely for its failure.  I don’t point this out to be smug, since I make mistakes all the time.  However, we’d be wise to not rely on its accuracy too much.

After playing around with Chat GPT for an hour, I think it’s safe to say that the promise of AI seems limitless, as do the risks.  While Chat GPT is all written words, there are plenty of photo and video “deep fakes” that are going to become a real problem in the near future.  People have already been terribly deceived on all kinds of things these days…imagine if the face of a person you trusted was faked into making it seem like he was doing something wrong.  Maybe like Tom Hanks kicking a puppy.  

I made the photo at the top of this post with AI…it’s accurate (“make a picture of a shark eating a pickle”), but it’s a little off, don’t you think?

 

So what are we to think about AI?

Like any brand-new technology, we don’t really know how it’s going to be used in the future, do we?  For instance, I’m old enough to remember that the first VCRs were advertised as machines you could use to watch one TV show while recording another on a different channel…which seemed like a real benefit since there were only six stations and there was no DVR.  Instead, the VCR turned into a movie-watching device, creating whole new industries around the idea you should be able to watch what you wanted when you wanted.

Remember the humble cell phone?  When it came out it was advertised as, well, a phone.  You could use it to talk to other people!  Silly us.  "Cell phones" today are essentially portable computers with a phone embedded in there somewhere.

The internet was initially a great way to sit and type to other people in real-time in chat rooms.  Oh, and it was great for circulating funny joke lists.  Now there isn’t an electronic device manufactured anymore that isn’t reliant upon it.  My birdfeeder uses the internet and AI.  (I’m not kidding!)

“The Facebook” was invented to get college students rank other college students by attractiveness…which morphed into what we know as Social Media.  For better and worse, Social Media has become a category of human interaction never conceived of when Mark Zuckerberg was still pissing off girls at Harvard.

Then there was a thing called Facetime, where you could see another person while you talked to them from great distances.  Now a few decades later, we conduct two of every three of our client meetings on a form of this called Zoom. 

What is AI going to be?  Right now, it writes funny stories about pickle-loving sharks.  It seems both pretty useless and pretty harmless.  Nonetheless, in a few years’ time it will no longer be either of those things.

 

I’ll let Chat GPT have the last word on this:

ME:  “Hey Chat GPT, write a concluding sentence on the promise and peril of AI in the future.”

Chat GPT:  “The promise of artificial intelligence lies in its potential to revolutionize industries and improve quality of life through automation and advanced decision-making, yet its rapid advancement also raises concerns about job displacement, ethical implications, and unintended consequences such as bias and loss of privacy.”

ME:  🤔. Hmm.  Sounds like every other new technology.  Stay tuned.