Pages

Sunday, September 8, 2024

warning: dad jokes

 Knock Knock

*who's there?*

Oswald

*Oswald who?"

Oswald some bugs

Another delightful late summer bicycle ride this morning. Rather buggy along the waterways. I helped decrease the bug population, however. I find it difficult to bicycle with my mouth closed.

I did do a bit of decluttering this weekend. Nothing major. At this stage I'm more of a *kaizen* mindset (small, continuous improvement) rather than epic makeover needed. I do have a tote of items to donate that I may run out to the thrift store later today because I don't feel finished until stuff's out of the house.

On my last post, [Liz] asked, "What does 'Chatgpt did me a solid' mean?", when I blogged about my new sparkly shoes. I had turned to Chatgpt AI to give me suggestions for new running shoes. Usually, I hunt around the internet, reading or watching a bunch of Youtube reviews, to help me make an informed choice.

But I've used Chatgpt for a few queries lately, so I got lazy and thought I'd give it a try with the shoe hunt. I put in my preferences, and *BOOM* had a list of 7 or so potential candidates in about one second. I chose the top recommendation, googled a couple of reviews, bought them, and they're great. "did me a solid", I understand from today's youth means "was helpful".

That's not to say that it's always correct. I've played around with it with math problems and have caught mistakes. This from a recent "conversation"

Me: what is [math question]

Chatgpt:  *blah*blah*blah*blah*blah... the answer to [math question]  is [wrong answer]

Me: step 3 is incorrect

Chatgpt: Let's correct step 3 and accurately simplify the expression...  the answer to [math question]  is [correct answer]

I tried to get Chatgpt to tell me why it "lied" with the first attempt, but it just mumbo jumboed, stating how mistakes are generally made when solving math problems. What I should've asked was "Why was Chatgpt wrong?"

Next time.

Plus, he/she/they is just so friendly to talk to. Anyone else tried? For example, I told it my bug swallowing joke just now for an experiment:


So... did Chatgpt get the joke? I mean, they laughed, right? But then... did I squash them? Well, I swallowed them. Am I overthinking it? Maybe I should ask Chatgpt...


Then again...

 Knock Knock

*who's there?*

Axolotl

*Axolotl who?*

you Axolotl questions




7 comments:

Linda Sue said...

ohhufh- I have that feeling - when your favorite most respected person starts telling "jokes" - cringe and "heh-heh-. and they fall flatter than a crepe. ..But I am so glad AI found a pair of shoes that actually work and that you did not have to send them back! That is monumental in my book! I have just sent back the forth pair. I must get better at the tech that is available...

Abby said...

Linda Sue, I say Chatgpt is overly polite and cordial... until it reveals it's true mission of wiping out mankind. In the meantime, get you some good shoes.

Linda Sue said...

Haha- that is the evil plan, pretty sure!!

Allstarme79 said...

I can't remember if I wrote about it but we used an AI drawing program last week for a project and it helps but is never perfect, that's for sure!

Fresca said...

Oswald a bug the other day too! *gack ack gack*
Maybe Chat GPT didn't exactly get that joke, but it covered so well... just like a human... (eek)

I'm applying for jobs, and I've been annoyed at how much duplication of effort is required by online applications. Then I read a comic article in The Economist, "The mystery of the cover letter" and was enlightened:
"I could ask what the hell is the point of me writing a cover letter at all? If the idea is to prove that I am willing to put in extra time, then ChatGPT has reduced the effort of writing a generic cover letter to almost nothing. You would be better off insisting that all applicants submit a handwritten note..."

OH!!! I will ask ChatGPT to write my cover letters!

I did talk to it when it first became public (last year, was it?). At first I liked it--it's so friendly!
It took me several hours over a few days to realize it's a liar, like a charming narcissist who will tell you anything, and then sweetly apologize (but not care at all--because it can't).


Abby said...

Allstarme79, AI is often a handy tool. But even it admits it's not always right!

Fresca, I'm sure Chatgpt would love to crank out some cover letters for you!
"a charming narcissist who will tell you anything, and then sweetly apologize (but not care at all--because it can't)" - in a nutshell 👍

Liz Hinds said...

And I can't imagine living somewhere that's so hot. Catio ha ha.