All Episodes

February 12, 2025 21 mins

On the Wednesday February 12, 2025 edition of The Armstrong & Getty One More Thing Podcast...

  • We dip our toes in some of the new Ai features on our smart phones! 

 

Stupid Should Hurt: https://www.armstrongandgetty.com/

See omnystudio.com/listener for privacy information.

Mark as Played
Transcript

Episode Transcript

Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
I didn't say, buy me a race horse. Oh no,
it's one more thing. I'm strong and getty.

Speaker 2 (00:10):
We're going to talk to a little artificial intelligence. And I
am god as much as I read about it and
I'm fascinated by it. You'd think I would be a
guy who is like paying for the premium chat GPT
and using it and trying it out, But I'm not so.
My first dipping the toe into AI has been what
has showed up on my iPhone in the last couple

(00:31):
of weeks with their new AI. The summaries of the
emails and texts.

Speaker 1 (00:35):
I love.

Speaker 2 (00:36):
It's really quite amazing. So I get a long text
from somebody and then just right there in the line,
is you know, just a few word summary of it?

Speaker 3 (00:44):
Huge fan of it?

Speaker 2 (00:45):
Oh, it's I love it. It's amazing. And then last
night I was I don't know if this always happens
or I just noticed it. So I got a really
long text from somebody about a complicated situation they're dealing with,
and I don't know how I feel about this, AI
suggested my reply. Oh, and I thought I'll come up
with my own replies.

Speaker 1 (01:06):
If that's okay, How must you pull yourself up by
your boat straps.

Speaker 2 (01:10):
Yeah, it was something like that.

Speaker 4 (01:12):
I had one the other day where I sent somewhat
of a long message and a little thing popped up
that said, would you like to correct your grammar?

Speaker 2 (01:19):
Like, no, don't tell me. I don't minding a book.
I don't mind hints on that. But this was like
a what take I should have on someone's problem? As
I replied to them, and I thought, that's that's not
your stay in your lane. AI.

Speaker 1 (01:38):
Wow, that's bold.

Speaker 2 (01:40):
Yeah, that's what I thought. I thought. I'll be damned well.

Speaker 1 (01:42):
I mean if if you described to me, Hey, if
it was crazy. A friend of mine called me last
night and said this, and I jumped in with, well,
you should tell him that he needs to you'd be like, dude, dude,
I'm just telling you about it. I didn't ask you
to like take it over.

Speaker 2 (01:56):
Right, And but so that on the other end of it,
what you've gotta be ready for is you could send
some long heartfelt text to somebody about a breakup or
something problem you're having with a job, and the reply
you get might be your lazy buddy just hit the suggestion.
I think you gotta look for a new job, probably
not gonna work out there or whatever.

Speaker 1 (02:17):
AI says, other fish in the sea, and then.

Speaker 4 (02:21):
They come back to you. Hey, Mike, thanks so much
for that advice.

Speaker 2 (02:25):
You're a real pal. You're always in a year I
can lean on. I appreciate that.

Speaker 1 (02:31):
Yeah, I'm starting to play with it a little bit
and enjoying it. I'll tell you this, friends, and this
is big fun if you go into image playground now on.
In messages, Jack doesn't use emojis because he lacks.

Speaker 2 (02:43):
The capacity for joy, but yet to send my first emoji.

Speaker 1 (02:47):
So, but you can now create AI images of yourself
in various moods and situations and that sort of thing.
And they are as my buddy Brian says, they're they're
quite realistic, they make you thirty percent better looking than
you are, cool, they're very amusing, and they will guaranteed

(03:09):
creep out your spouse or partner. Yes, it's there seems
to be one hundred percent correlation between sending it to
your loved one and then being creeped out by it,
which is hard to describe why exactly. Maybe it's that
what's the It's not Okham's rais or the Infinite Valley

(03:31):
or the Uncanny Valley.

Speaker 3 (03:32):
Uncanny Valley.

Speaker 1 (03:33):
Yeah, something that is almost right is disturbing to us
as humans for some reason.

Speaker 3 (03:38):
This isn't almost almost right.

Speaker 4 (03:39):
But the next step up in the playground is you
can make use that same technology and turn yourself into
an emoji that you can attach to texts. I just
sent you the one that it created of me. I
sent I put in a picture of myself and said
talking into a microphone. Yeah, so now I can throw
that little chick into texts. Oh, but it's it's problematic.
How much fun this this new AI is on these phones?

Speaker 1 (04:02):
Oh yeah, here, stand by Jack. This is a very odd,
very odd here's uh, here's roughly what I sent yesterday.

Speaker 4 (04:12):
Okay that and it's amazing how fast it does it too,
when you put the photo or whatever information in you need.

Speaker 3 (04:19):
It's it's a matter of seconds.

Speaker 1 (04:21):
Yeah, all right, go sending it to the group. Dig
this this is the and birthdepw Wow, how do you
do this? Guys? Where do you find this sent image playground?

Speaker 2 (04:38):
It does make you thir weirder, looking better because it's weird.

Speaker 1 (04:43):
Better looking, But it's it's all putting to Judy.

Speaker 2 (04:46):
I'll tell you that I'm not hitting on you, but
you're better looking than that.

Speaker 1 (04:51):
I don't know.

Speaker 3 (04:51):
That's a weirder teeth than that, for sure.

Speaker 2 (04:55):
Teeth. That guy could use some white ners.

Speaker 1 (04:58):
Yeah, mix it a little fluoride. Oh my god, Yeah,
I've got I got like summer teeth. Like it's like,
is this did I accidentally say and make me British?

Speaker 2 (05:09):
Yeah?

Speaker 4 (05:11):
There should be a piece of hay tucked in between
those two front ones right there.

Speaker 2 (05:14):
Or make me homeless room and they still got a
few bugs to work out. I was messing with it
a little bit the other day when it first landed
on my phone, and I just did a I just
voice texted a I want to bear in a cowboy hat,
and it had and he gave me a baron a
cowboy hat and then dancing, and then it was a
baron a cowboy hat dancing, and just I tell it.

Speaker 1 (05:34):
I want my wife and a cowboy hat.

Speaker 2 (05:37):
I've seen plenty of that stuff. Yeah, you can do
that all day long, do it, but it's just it's amazing.
And if I would have loved that so much as
a kid, oh my god, it would have blown my mind.

Speaker 3 (05:45):
So for everybody wondering and Michael, I know you just asked.

Speaker 4 (05:48):
If you go to your settings on your iPhone and
scroll down, it says Apple Intelligence and Siri, and you
have to turn that on.

Speaker 5 (05:55):
It.

Speaker 4 (05:55):
I'll go through a download and upgrade process. But then
all of a sudden, playground and whatnot will appe.

Speaker 1 (06:00):
And it's the little plus sign next to the field
where you write a text where you can add photographs
or whatever picture you have to.

Speaker 5 (06:07):
Have the newest iPhone or fifteen and up. Okay, good,
I do got fifteen now, so all right?

Speaker 1 (06:12):
Yeah, so anyway, isn't it oddly off putting?

Speaker 2 (06:17):
I'm trying to fill out figure out how you prompt
it to give you the AI suggestion. It looks like
it has to be a recent text you got, but
I but I got one today, like somebody texted me
something and AI says I should reply with I get that.

Speaker 1 (06:36):
I don't.

Speaker 2 (06:36):
I don't need your help on that.

Speaker 1 (06:37):
All right, let me see if it will it work
in a group text.

Speaker 2 (06:44):
I don't know. I just noticed it last night, and
the one it gave me last night was on this
really complicated friend of mine's life situation, and it said
it gave me some suggestions and I thought, this is
really not appropriate. Everybody's typing in their phone. I know,
I got a dance a cowboy hat, and I like,

(07:06):
how if you say a lot like my wife, if
you want a dancing bear in a cowboy hat, it'll
give you like five versions, fat bear, thin bear, cowboy
hat and vest in boots or cowboy hat and just
you know, a six gun or whatever.

Speaker 4 (07:19):
I for some reason asked it to make a pickle
working out. So I have a pickle that's lifting a
bar bell.

Speaker 3 (07:24):
There you go, that's a fun one.

Speaker 2 (07:26):
Makes sense.

Speaker 1 (07:26):
Yeah, okay, let me send you this Jack and see
what it summarizes and whether it suggests a boy a reply?
What Jack read? That's real quick?

Speaker 2 (07:39):
Yeah, so it's not suggesting a reply. And I don't
know why. I don't know why it sometimes doesn't sometimes doesn't.

Speaker 1 (07:46):
Did it have a summary or was it too short
to need it?

Speaker 2 (07:49):
I think it's too sure.

Speaker 1 (07:50):
I wrote you were an idiot and held me back
my entire career. I quit, lose my number.

Speaker 2 (07:55):
Lose my number. That pretty funny.

Speaker 1 (08:04):
So the whole I didn't tell you to buy me
a racehorse introduction was about going into Jeff Fowler's column
in the Washington Post recently great tech writer, but he
was trying to assess the capability of AI at this point,
using a couple of different systems, and he had He

(08:24):
gave it a couple of missions. One was fairly successful.
He tested Operator, which is what company is that from?
I can't remember, it doesn't matter, Google it. But he
tasked Operator with interacting with my internet service provider, go
into my Comcast Xfinity account and see if you can

(08:47):
find me a less expensive plan. Operator applied, all right,
but thirty seconds later it stopped. It needed my log
into the Exfinity website. The problem is Operator didn't know
much about the nitty gritty details of your life, but
it needs your data to actually be helpful, so it
often pauses and asks for help, and it tries to
shield your privacy, which also means that often it stops

(09:09):
and says, hey, can I use this or can you
give me this whatever? But then this is where I
got more interesting. Once Operator was logged onto my Comcast account,
it took about two minutes to do something incredible. It
found a way to save me money, well sort of,
it said, it found an alternate internet plan for thirteen
dollars per month. That seemed awfully low because I currently

(09:29):
pay sixty eight.

Speaker 2 (09:30):
I was going to say, I think Lend's like eighty.

Speaker 1 (09:32):
Yeah, So I inspected it's browser window and saw a
Comcast was really saying this plan would be minus thirteen
dollars compared to my current plan, not thirteen dollars. Okay,
it missed the minus sign. Another big question is whether
AI can. About AI is whether it can understand enough
about the real world or even just the web to
operate in it. Repeatedly in my tests, I saw that

(09:54):
operator could misinterpret what it saw in its browser. Now
this part it is interesting. In this case, Operator redeemed
itself after I asked it to spell out the full prices,
including taxes and fees. It gave the right total and
did one better. It read all the fine print and
found that this deal would go up by sixteen dollars

(10:16):
after an introductory period, making it a bad deal.

Speaker 2 (10:20):
That is helpful.

Speaker 4 (10:21):
I didn't even think about having AI read the you know,
the terms and.

Speaker 2 (10:26):
Condition summarize the ten pages where it's going to watch
every keystroke or whatever it says.

Speaker 1 (10:32):
Right, and at this point it might miss the salient points,
the really important stuff. But the fact that it's moving
in that direction. Hey, read this legal lease and tell
me the parts I need to know would be super helpful, obviously,
But then the failure that was a success, he said.
The failure was he said, find the cheapest set of

(10:53):
a dozen eggs I can have delivered. Then I gave
it my address to conduct its search. It needed my
logins for grocery delivery services. Blah blah blah. I didn't
think about it at the moment, but doing so also
gave Operator access to the credit cards I'd saved with
those services.

Speaker 2 (11:08):
About to say, I'm not willing to give aim at
credit card and see, I wonder how it does with this.

Speaker 1 (11:14):
Initially, Operator found some five ninety nine a dozen eggs
on a site called Mercado, but noticed there was a
twenty dollars minimum order requirement. Good job told I told
it that it could add additional eggs to check the
final price, but it decided to switch it's onto Instacart.
Then Operator went quiet as it clicked around, and I
walked away from my computer. A few minutes later, I
got an alert from the credit card app on my phone.
I had just made a purchase on Instacart. What happened?

(11:36):
How do I stop it? I gasped, Was there any
chance the AI might go on a bigger shopping spree?
I hadn't told it to buy eggs, just find cheap ones.
I was able to reconstruct some of what had happened.
On the instacart website. Operator found a dozen large white
eggs for thirteen dollars and nineteen cents, more than double
the other site. For unclear reasons, it purchased these, added
a three dollar tip and a three dollars priority fee

(11:59):
on top of a seven dollars and ninety nine seven
ninety nine delivery fee, four dollars service pee fee, and
twenty five cent bag fee. Thankfully, at least operator climbed
and offered to sign up for an Instacart membership blah
blah blah, And they got the wrong number on that actually,
and it cost him thirty three dollars I think it
was for his eggs. Thirty one dollars on a dozen eggs.

Speaker 2 (12:23):
That's pretty interesting that it didn't do that and then
say would you like to purchase it just went ahead
and purchased them.

Speaker 4 (12:30):
And then you have to call the credit card company
and say sorry, that was AI that did that.

Speaker 3 (12:34):
It wasn't me forgery.

Speaker 1 (12:35):
You make some omelets?

Speaker 2 (12:37):
Yeah, I called the credit card company, said I can't
eat this many eggs? What should I do?

Speaker 1 (12:42):
Oh my gosh, So who did Jack? Did you do that? You?

Speaker 3 (12:45):
I did Jack?

Speaker 1 (12:46):
Oh? Oh my gosh. Have you looked at that yet? No,
don't worry. It doesn't totally make you look like a rube.

Speaker 2 (12:55):
Oh my god, Oh my god, that is that's not
thirty I hope that's not thirty percent better looking. I'll
just I'll just make that my my dating profile if
I ever go online.

Speaker 6 (13:08):
Here you go, Oh my god, that's brutal. Oh my god,
my kids are going to you said, so you got
to plunk out this theme from Deliverance on your jet.

Speaker 5 (13:21):
Yeah, that's supposed to be a picture of me. It
didn't quite come out right at all.

Speaker 2 (13:24):
A second, how come you get to look like that?

Speaker 1 (13:27):
You know?

Speaker 5 (13:27):
All I put was dark hair, five foot seven, and
it said, you know, gifts it and I put leather pants.

Speaker 2 (13:32):
Why does AI make you look like that? But makes
me look like this?

Speaker 4 (13:37):
Michael gave a description of himself. I actually put a
picture of you in there, and it pulled from a photo.

Speaker 2 (13:43):
No, Michael looks like you know, Timothy Shallomet at his best,
I mean, hot, angrogynous model type. I look like somebody
you would run from or donate to. Right, I was
old it, this is crazy angry look. I gotta send

(14:05):
that to a few people and see how they react.

Speaker 1 (14:08):
You know what, It's funny. I gave it a clearly angry,
disapproving look, and it at most it made me look
oddly manic, But the rest are pleasantly contemplative. It's like
Apple won't let me be angry, just it.

Speaker 3 (14:24):
Doesn't want you to be likes happy Joe.

Speaker 2 (14:28):
I'm gonna I'm gonna spend more time paying attention to
Oh my god, I'm going to spend more time paying
attention to the suggested replies to text me now that
I'm getting them, because I went to one last night
my son. I texted him something and he said, it's
really late. I'm getting ready to go to bed. And

(14:48):
then AI suggested I say okay, good night, which is
you know, reasonable, I guess just to avoid having to
type it out or say it.

Speaker 1 (14:57):
You know, I tell you what though, you use any
keyboard shortcuts?

Speaker 2 (15:02):
No, I'm familiar with the term, but I don't use them.

Speaker 1 (15:05):
Yeah, I do, for partly for when I'm forwarding articles
to myself at mailbag at Armstrong and geddy dot com.
It's so cumbersome to type out and if it doesn't autofill,
I just do the first five letters and it autofills
for me, no matter what, whether the site's working properly
or not. And actually there's a funny story connected to this.
My son discovered this years ago. You could do that,

(15:28):
and for whatever angsty teenage reason at the time or
very early twenties, he thought it was just so dumb
and cliched when people would text lol, And so he
and we shared an Apple account in a way that
doesn't exist anymore. But so I would type lol. I
remember talking about this on the air, and it would

(15:50):
instead of that, it would auto fill a phrase that was,
why don't you blank my face because I'm a blanking blank.

Speaker 3 (16:01):
Okay, I remember this feature?

Speaker 1 (16:03):
Yes, yes, that's hilarious. Oh it was, And I typed
that and I'm like, what the f Wait a minute, honey, honey.
I went to gudy and look what my phone just did.
What the hell is this? Oh that's brilliant, And it
was an unrepeatable phrase.

Speaker 4 (16:21):
I have to turn that off on my phone, I
think for a couple of reasons. Because I tried to
say I was d D the other night to a
friend because I was designated driver, and for some reason,
my phone thinks d D means sorry, I'm driving, I'll
text you later, and it wouldn't let me just say
d D. So yeah, it's a little a little frustrating.

Speaker 1 (16:40):
I don't know, but anyway, I'm sorry. To get back
to Jack's point. Judy, who is my wife, who's a
very busy person and does various things around town, musical
things and craft things and whatever. She always texts me
when she's on her way home, and you know, it's
usually it's always on my way exclamation point, and you know,

(17:03):
I reply with a hard or whatever, a great scene,
a few or whatever. But she revealed that she has
programmed her phone so all she has to do is
do like O and space M and it finishes the
message for her. So she's not actually typing out on
my way exclamation point, her phone is doing it for her.

(17:24):
And I find myself resenting that just a little bit.

Speaker 2 (17:28):
That's interesting. That's what I was wondering about. These replies.
I don't have any problem with saying to my son
good night, obviously, but the computer suggesting it and me
just agreeing to it.

Speaker 1 (17:43):
Having the computer wish your son good night on your
behalf certain.

Speaker 2 (17:48):
Yeah, it seems a little weird.

Speaker 1 (17:50):
Does seem a little Victorian. The nanny interacts with the
child ninety eight percent of the time, and on Sundays
I will go for a walk with the boy and
then send it back to boarding. It just seems a
little why is there a little man.

Speaker 2 (18:02):
Here to right?

Speaker 1 (18:03):
Right?

Speaker 2 (18:04):
And the example I had of somebody sending me a
long heartfelt text about something I never remembered it was
last night, and then the AI suggesting a reply, I
thought that would be really cheapening the seriousness of this conversation.
Not that they would know, but it's.

Speaker 4 (18:18):
I don't know if it's something that you're already going
to say and the phone assists you, like how Judy
does the on M and it says on my way,
I mean she was already going to type that you
Jack were already going to say good night.

Speaker 2 (18:32):
It's not like But the other example I have it
was something like, I think you're on the right track.
I hope that works out for you after a really
long heartfelt text about some difficult situation. And that's just
that ain't right.

Speaker 1 (18:44):
That that ain't right. So final note that last picture
I sent you. I wish I could send the original picture,
but I can't figure out that was me like angrily
grimacing at the phone, and Apple turned it into a
you know, cheer alert picture.

Speaker 4 (19:01):
I think it's very Stepford wives. First of all, this
aged you for some reason. I don't know why I
did that.

Speaker 3 (19:08):
This photo. Hell no, you're not.

Speaker 4 (19:11):
But the second part of it is in that prompt
you can tell it the mood you want the photo
to be in. So m so you can grimace and
then tell Apple, hey, make me look like I'm like
I'm pissed, and it should do it for you.

Speaker 1 (19:26):
And then you can add cowboy and whatever else a
series of prompts for it to.

Speaker 2 (19:31):
But one thing Apple needs to figure out what these
made up emojis of you is. They don't flossing is exists.
Flossing exists, toothpaste exists. Let's teeth, Yeah, the teeth are bad.
What is going on there?

Speaker 1 (19:48):
It's sounds like like I got newscaster teeth or anything.

Speaker 2 (19:51):
But they're okay, but they're not brown with those big
gaps in between. I mean, what the heck is going
on there?

Speaker 1 (19:57):
All right, one more one more effort? Make Apple make
me look angry? Right, there's my grimacing. I'm even got
like one eye closed like I'm Popeye and then angry.
Will Tim Cook permit me to feel natural human emotions

(20:20):
are not? Oh hey, that's pretty good. Wow, that's like
call security. Okay, Tim, I retract the criticism. I just
wasn't good at it.

Speaker 5 (20:33):
That's a good one, Katie.

Speaker 2 (20:36):
Yeah, that actually looks like you. Katie. What was the
prompt that got a cartoon drawing of you that actually
looks like you? I just a picture of it, used
a photo of you?

Speaker 1 (20:45):
Okay, Well you have the advantage of being super cute.

Speaker 2 (20:48):
Yeah that makes it easier.

Speaker 3 (20:50):
Yeah, you guys, it's two weeks, Joe.

Speaker 1 (20:52):
This last one of you is great. Yeah, that's that's
you know you live with me. You're gonna see that face.
I'm gonna use it. What I'll tweet that right now
our ex account so you can see it.

Speaker 5 (21:08):
I'm gonna use AI to win arguments with my wife.
I'll just look at it and wait for the response,
and this is what I'll tell her exactly.

Speaker 1 (21:15):
You take over here. Yeah, well, I guess that's it.
Advertise With Us

Hosts And Creators

Joe Getty

Joe Getty

Jack Armstrong

Jack Armstrong

Popular Podcasts

Crime Junkie

Crime Junkie

Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.

24/7 News: The Latest

24/7 News: The Latest

The latest news in 4 minutes updated every hour, every day.

Stuff You Should Know

Stuff You Should Know

If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.

Music, radio and podcasts, all free. Listen online or download the iHeart App.

Connect

© 2025 iHeartMedia, Inc.