January 28, 2025 • 24 mins
Gary and Shannon begin the show with a deep dive in China’s DeepSeek AI app and President Trump saying, ‘this should be a wake-up call.’
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
This is Gary and Shannon and you're listening to KFI
AM six forty, the Gary and Shannon Show on demand
on the iHeartRadio app. That's just said, that's not sad,
it's a data think about her and to celebrate her life.
Speaker 2 (00:14):
But I didn't think about it until I just looked
at the rundown. Oh.
Speaker 1 (00:19):
I mean, well, you haven't thought about twenty eight. I
haven't thought about the between twenty eight, right. It'd be
different if you saw it earlier and just forgot.
Speaker 2 (00:27):
We're finally doing a burial. If I told you that,
Oh good, when is that? That's gonna be next month?
That's nice? Hey, what's the rush?
Speaker 3 (00:36):
I know?
Speaker 2 (00:36):
And that seems to be the families. That's that's the
way we've always done it. Come on, we don't have
any religious obligations to bury people within twenty four hours,
so we wait, you know, fourteen months or in some
cases two years.
Speaker 1 (00:49):
Are why did you wait so long? It's not like
you're planning a party or anything.
Speaker 2 (00:55):
No, and it's just going to be it's just going
to be family. Yeah, So I don't know, I don't
know why. How do you make those decisions? I mean,
that's first of all we had to figure out where
because there's two family there's kind of two family cemetery things,
and one is mom's generally mom's side, the other one
is generally Dad's side.
Speaker 1 (01:14):
I can see if this was a you thing where
you're kind of like, eh, whatever, But your sisters seem.
Speaker 2 (01:18):
Kind of type A to me. Yeah, I mean there
is that.
Speaker 1 (01:22):
So that's kind of different to me that they would
wait so long. Yeah, well, but there's been a lot
of there's been a lot going on, a lot, there's
been a lot going on, so it's probably fine. It's
actually kind of nice that you can do it with
all the other crap to get done with the house
and the property and all that. Now that that's been
on that.
Speaker 2 (01:40):
True, Yeah, that is true. So anyway, not to start
on a negative note.
Speaker 1 (01:44):
Wow, well you want to do more dead stuff?
Speaker 2 (01:46):
We could. I found out that if if a bible
becomes damaged, the way that you dispose of it is
however you want to dispose of it.
Speaker 1 (01:58):
Why would you dispose of it just because it was damaged.
Speaker 2 (02:00):
Well, there was a time in the most recent past,
I'd say within the last week that my dog shredded
a bible.
Speaker 1 (02:09):
Really, so it wasn't just the yoga blocks.
Speaker 2 (02:12):
It was not just the yoga blocks. In fact, this
was pre yoga block. He went after and just destroyed
a Bible that was sitting on the couch. What do
you think that means? I do not know, but I
did ask for pastoral advice. I texted a friend of
mine who was a pastor, and I asked him what,
first of all, what I should do with dog? And well,
(02:33):
clearly the dog is not redeemed. Put the dog down.
And then I said, well, how do I dispose of
a Bible? Is there like is there.
Speaker 1 (02:42):
Is there like some sort of turn it in?
Speaker 4 (02:44):
Well?
Speaker 2 (02:45):
Ak, is it like an American flag? Like there's there's
procedures and policies and physicians. And he said, yeah, put
it in the recycle bin. Was a jump dum.
Speaker 1 (02:58):
It's a book. So it was sitting on the couch.
Speaker 2 (03:02):
Huh. That does that mean that you regularly read it?
My wife was reading it? Oh good, Yeah, that's good.
Speaker 1 (03:08):
At least there's one of you in that house that
is God fearing and.
Speaker 2 (03:13):
I have a Bible. Yeah, but you don't read it.
Speaker 1 (03:17):
In the other room behind behind the gate. Why is
the Bible behind the.
Speaker 2 (03:22):
Gate so that the dog doesn't chew it up. When's
the last time you picked up your Bible? You mean
the book or in my phone? I have an app
on my phone.
Speaker 5 (03:30):
Oh you do?
Speaker 2 (03:31):
Okay, I bring up right there, Matthew fifteen is what
it was. Oh okay. Some ps two Psalm two, Verse one,
Why do the nations rage and the people's plot in vain?
The kings of the earth set themselves and the rulers
take counsel together against the Lord and against his anointed.
Speaker 1 (03:53):
That's exactly what I was thinking of when I was
thinking about this Ai fight between well, China and everybody else.
This because the kings of the world of Ai, they're
doing what the Psalm says.
Speaker 2 (04:05):
Well, and they're doing a kind of behind curtains. I
don't want to say behind closed doors, because there's some
information that we know, but there's so little that we
know about the generation of deep seek Ai. Why I
mean this it is, it's not just chess.
Speaker 1 (04:24):
Here's China playing chess with our checkers. You know, Biden
going out and you know, banning companies here in the
United States from exporting AI chips to China was kind
of like, we're not going to give you our super powerful,
super expensive chips, and China's like f us, no f u,
We'll do it on our own for cheaper. We don't
(04:46):
need your chips. We look like fools in this, like
we thought that everyone needed our super expensive chips to
go and make their AI advances and that's not the case.
We look like the dumb dums on the global state
right now. And Trump is right that this should be
a wake up call.
Speaker 2 (05:05):
And then all it takes is a Chinese mcguiver to go,
oh well, I don't have the greatest, the fastest, the
biggest chips. Give me a ballpoint pen, ballpoint pen, some
duct tape, and then all.
Speaker 1 (05:17):
The trans transmission it used to be a transmission. Tranny
was a word for a transmission before I do any other,
meaning let's.
Speaker 2 (05:25):
Bring that back. So yeah, there's a lot that's that's
going on. Also, I have a question that I just
thought of, who is using this on a regular basis?
Who uses chat GPT?
Speaker 1 (05:38):
I think ki, I think it's gonna be everyone's using
it in the future.
Speaker 2 (05:43):
Right, Well, I mean because to me. Okay, so there's
a couple there's a couple of things. This morning I
realized Wall Street Journal now summarizes their articles. There's three
bullet points, and I think a lot of news organizations
do this now that I recognize it. Three bullet points
at the top of a long article. You can read
those three bullet points and get an idea of what
the article is. And there's a little thing in the
(06:04):
corner that says, what's this. I clicked on that this morning,
and it says, this is an AI generated summary of
the article that you're about to read. Well, no one's
going to read it if AI already just summed it
up in three bullet points. That's one thing. The other
thing is Google, when you search for something now automatically
gives you the AI version compendium of everything it's searched for.
(06:29):
Apple is trying to get you to update all of
your iOS and operating system stuff with its AI version.
It jammed down your throat.
Speaker 1 (06:39):
I was shocked, and not so shocked with the quickness
that I adopted Google's AI search results. When they first
started doing that, I said, what the F is this?
Screw this, I'm not paying attention to that. I would
scroll down, I would click on the various different things
that popped up to reach the summation that the AI
(07:00):
bullet points would have given me right off the bat.
So now, I mean it was probably a matter of
days where I was putting up a strong front of
I'm not using AI, I'm not using Google's AI, the
AI bullet points are feeding me, I'm not using that.
It was probably a matter of like eight to twelve
days between that and me gobbling it all up.
Speaker 2 (07:22):
Can't stop it because it's easy and we're lazy people, Okay,
but I mean, who's For example, this deepseak ai has
skyrocketed over the last three or four days to the
top of Apple's App Store downloads. Why what are people
just curious about it? Because when chat gpt came out,
(07:42):
people were like, make a poem about a monkey, hump
and a deer, and.
Speaker 1 (07:47):
Make a poem about penis, and people went with that
because it was a curiosity, right. I think a lot
of people use this for work related purposes.
Speaker 2 (07:57):
Oh yeah, school, I will say this. I did ask
chat gpt to write a radio play for our holiday
show and it was awful.
Speaker 1 (08:10):
Okay, so here's something that you should know. We don't
have real jobs, right, so we can't fathom how people
would use this in a practical and time saving work
smarter not harder manner.
Speaker 2 (08:23):
I'm hoping this is the I want to learn. I
want if you use AI on a regular basis, whether
it's Chat, GPT or deep seek or whatever it is,
if you use it on a regular basis, I'd like
to know how you can leave us a message on
the talkback feature on the iHeart app. Wall Street is
some you know, turning to some sense of normalcy after
(08:46):
what happened yesterday when, for example, Nvideo, one of the
big AI chip makers, lost more than five hundred billion
dollars in market value in one day, five hundred bit.
It had five hundred billion to lose, and it lost
five hundred billion dollars.
Speaker 1 (09:04):
A couple bits of feedback. Look at David Muir's muscles,
My goodness, is that necessary to put those out?
Speaker 2 (09:13):
It's a hot day in that helicopter. That's a G
I Joe commercial, right there, wasn't it? That's exactly what
it was. Okay.
Speaker 1 (09:22):
We've gotten a few bits of feedback with your dog
eating the Bible.
Speaker 2 (09:26):
One bit of feedback.
Speaker 1 (09:27):
Is Gary's dog is metal as f okay. The other
is that you should knock out five hail mary's that
you did not protect the Bible, that the dog was
doing dog things.
Speaker 2 (09:39):
But it's up to you to protect the Bible. So
protect the Bible.
Speaker 1 (09:42):
So you're good for five hail marys, which I think
is low. I mean I would go with, yeah, protect
the Bible. I don't do hail mary's. That's n what's
a Christian prayer? Just like our father? You do that one,
but you do the our Father right, everyone does, sire?
Speaker 2 (09:56):
Yeah, everybody kind of does that one. All right, how
about give me five ur our fathers will call it again.
That's not how it works. It is no, no, no,
you don't pay a toll to get across the bridge.
Speaker 1 (10:06):
Well no, you've already crossed the bridge and then burned
it down, so you now have to pay the damage
fee for the Bible.
Speaker 2 (10:15):
Yeah it is good morning, Hey, Ary.
Speaker 6 (10:20):
I know you, you know your upbringing was a little
unorthodoxed and you know, some weird stuff, But I think
it's all cool. But the fact that you guys aren't
going to bury your mom next to your dad, I
find that kind of strange. I think that's a no brainer.
Speaker 2 (10:35):
You know if I'm wrong, aren't you?
Speaker 6 (10:37):
But I can't see any other place. I think that's
the most appropriate thing to do.
Speaker 2 (10:42):
Feel very strongly about people. He doesn't know that. Well.
First of all, f you, you don't get to tell
me where they're gonna They get to decide where they're
going to be buried. And yes, they're going to be
buried together. That's such a weird.
Speaker 1 (10:57):
I'm pretty sure my husband has it somewhere in writing
to not be buried next time.
Speaker 2 (11:04):
Don't care what you do in life separated and give
me some f and pece around here. That's great. Good morning, Gary.
Speaker 7 (11:13):
Hey, did you ever get rid of that squatter in
your parents' house?
Speaker 1 (11:16):
Yes, yes, that was That was one of the things
you had to take care of before the burial.
Speaker 2 (11:21):
They would not have wanted us to have that around.
Speaker 1 (11:24):
Hi, Gary, I love you both.
Speaker 8 (11:26):
Hey. I am a nurse case manager for an insurance company.
I work in their workers' comp division, and we use
AI every day to kind of as a collaborative partner
for medical reports, looking at diagnostics and anticipating what type
of care a patient or an injured employee might need.
Speaker 2 (11:49):
Okay, talk to you so well, that's good. I mean
that makes sense. We're going to talk in a couple
of minutes here about what Deep seek Ai is, why
it caused such an earthquake on Wall Street yesterday. But
this was a comment from President Trump just a few
days ago.
Speaker 3 (12:03):
The release of Deepseek Ai from a Chinese company should
be a wake up call for our industries that we
need to be laser focused on competing to win. Because
we have the greatest scientists in the world. We always
have the ideas we're always first.
Speaker 2 (12:18):
We may always be first, But China did it cheaper.
Speaker 1 (12:22):
It's time for us to stop thinking that way, because
it's just not the case anymore.
Speaker 2 (12:27):
I don't think it's the case anymore.
Speaker 1 (12:28):
If this is real, if China jumps out and says
we can do it for cheaper, or we thought that
we held the key to the expensive computer chips that
make this all possible. I mean, as I mentioned, the
Biden administration band companies from sending these powerful AI chips
to China. So we must have thought that we were
(12:49):
the big kid on the block.
Speaker 2 (12:51):
We're not. If they're able to do this, you raise
a great question though, if this is real. Remember this
is a China needs product, and with that comes a
certain amount of mystery about its background. About its origins,
about how it was actually created, who funds it, and
what kind of information is it going to take when
(13:12):
you use it? Where does that information go? Yes, echoes
of TikTok.
Speaker 1 (13:16):
They have shown some of their math. They did publish
a technical paper last week about exactly how they were
able to do this PS. They're making it available for
other people to get in there and tweak it or
change it.
Speaker 2 (13:30):
And here's why. They want as much.
Speaker 1 (13:32):
Information about us or you or the smartest of the
smart that they can get. And if you get into
their system and you start tinkering around, guess what they're
tinkering with all of your stuff. It looks like most Americans,
more Americans, I should say, are working two jobs. Rising
caused a strong side market. Do you have a side gig?
A side market? Are you selling feet picks? What's going on?
(13:54):
There's no judgment here.
Speaker 2 (13:56):
Yeah, they would be total judgment. No, there wouldn't.
Speaker 1 (13:58):
Hey, if your feed are nice and have to send
feet picks, I mean I would have to pay people
to look at my feet.
Speaker 2 (14:04):
They're so gross. Be careful what you'd say. People are
into gross feet people lid for every pot. Wow, that's
really bad. I never thought about that. An eyeball for
every corn.
Speaker 1 (14:19):
Also, at the top of the hour, we're going to
tak an eyeball for every corn.
Speaker 2 (14:23):
Yeah, the corns on your feet.
Speaker 1 (14:24):
Oh oh, I don't have corns bunions, No, but look
co fungus. Look, oh god, they're not great.
Speaker 2 (14:36):
Ah here, wipe off the You're gonna throw a bare
foot up there. Also at the top of the hour,
the creek fire. If you remember that from twenty seventeen,
so cal Edison said, hey, it wasn't our power lines.
The FEDS think so cal Edison may be lying about that.
(14:58):
Of course you think of that in the context of
the eating fire, and now people pointing their fingers towards
the utility for that one as well. So we'll talk
about that coming up.
Speaker 7 (15:05):
A game changing move that does not come from Open Ai,
Google or Meta. There is a new model that has
all of the valley buzzing, but.
Speaker 2 (15:14):
From a Chinese lab called deep Sea.
Speaker 9 (15:17):
It's opened a lot of eyes of like what is
actually happening in ai in China.
Speaker 1 (15:23):
Deep Seek is a Chinese AI startup. It was founded
in twenty twenty three by Langwin Fang. He's the co
founder of a hedge fund high flyer. High Flyer uses
AI algorithms to predict market swings, so you can see
why he's got stock in the game, right. If you're
going to use AI to protict market swings and make
an s ton of money, not for yourself alone, but
(15:46):
for everybody else, you want your AI to be tip top.
Speaker 2 (15:50):
You want it to be accessible and cheap. So that's
what he set out to do.
Speaker 1 (15:55):
They claim that it's AI software is competitive with many
main made by US tech giants like open Ai and
Google at a fraction of the cost. Last week, Deep
Seek did publish the technical infrastructure, which talks about the
AI model behind the app. It claims cost savings from
(16:18):
using far fewer and less advanced computer chips than advanced
AI projects usually do, like the ones we do. It's
punitively expensive here in the United States because we've thought
all along that we need to use these advanced computer
chips that are more expensive.
Speaker 2 (16:34):
Turns out you don't. That is the biggest issue.
Speaker 1 (16:38):
That's the biggest gift though to us as well. If
we can harness that and create our own AI infrastructure cheaper,
it's a win for everybody, isn't it.
Speaker 2 (16:49):
Chaitan Puta Gunta is a general partner at Benchmark, and
I heard him on CNN explaining, we still don't know
a lot. I mean, as much as we do know
about deep Seek, there's plenty other questions.
Speaker 9 (17:00):
So did they actually assemble this talent, how did they
assemble all the hardware? How did they assemble the data
to do all this? We don't know, and it's never
been publicized. And hopefully we can learn that.
Speaker 2 (17:10):
Yeah, hopefully we can learn that. So here's a terrifying thing.
Alexander Wong is a founder of Scale Ai, and he
actually that company is helps AI systems learn. It kind
of puts together the information that AI then dives into
to learn from. And this is a terrifying aspect of
(17:34):
There are two parts of his answer here that are terrifying.
Speaker 7 (17:36):
Today we released Humanity's Last Exam, which is a new
evaluation or benchmark of AI models that we produced by
getting math, physics, biology, chemistry professors to provide the hardest
questions they could possibly imagine. Deep Seek, which is the
leading Chinese AI lab. Their model is actually the top
(17:58):
performing were roughly on par with the best American males.
Speaker 2 (18:02):
And again I mean that point of it's using it
with just a tiny percentage of the money that we've
put into AI with chips that we I mean, the
lower quality, slower chips that we have been thinking, Will
they never be able to put together an AI system
if they use those low quality chips? Can I can?
(18:22):
I just want to point something out here, he referred to.
He referred to something called Humanity's Last Exam. This is
a this is a test that they put forth for
these different AI models and the Center for AI Safety.
(18:43):
It's a test that they say as a groundbreaking AR
benchmark designed to test the limits of AI knowledge at
the frontiers of human expertise. And as he mentioned, they
just ask some of the smartest people in the world
come up with the absolute hardest questions that you can,
some of which may not act actually have answers that
we know, and see if AI can do anything with it.
(19:03):
I'm going to give you one example of the kind
of question that they ask AI to come up with
an answer for question one. Hummingbirds within a podiformis uniquely
have a bilaterally paired oval bone a sesamoid embedded in
the quadrilateral portion of the expanded curetiate upeneurosis of insertion
(19:28):
of M depressor kaudia. How many paired tendons are supported
by this sesamoid bone answer with a number twenty three.
I don't know if that's true. I don't know if
that I.
Speaker 1 (19:43):
Do deep Seek in terms of if it's safe, by
the way, that's very convoluted that question, like yay for
their AI if they can. They also ask about you
should know the answer.
Speaker 2 (19:53):
To biblical translations from a Hebrew in Greek mythology, who
was Jason's maternals, Jason's maternal great grand.
Speaker 1 (20:02):
Thoughts, So it's got a wide breadth of knowledge there.
Deep Seek, by the way, in terms of safety, can
collect and store a massive amount of personal information, including
any conversations you have, technical information about things like your device,
your Internet connection. This data can be accessed by the
Chinese government.
Speaker 2 (20:22):
Okay.
Speaker 1 (20:24):
The app also appears to be censoring certain information like
I don't know Tienneman Square and the protests there. Yeah,
right in the line of what China does now. Deep
Seek has released a version of its AI model for
others to use and modify. Outside developers can make it
more secure. Both more secure and less restricted by running
(20:45):
it locally on a powerful enough device. But yeah, the
fact that they've made it freely available for others to
download and modify means that they just want more people's
information in its clutches.
Speaker 6 (20:58):
Actually just USEPT in my car this morning.
Speaker 10 (21:02):
I'm on the way to go do some sales calls, and.
Speaker 6 (21:04):
I bounced some ideas off of chat GPT as far
as the message I want to get across my clients.
Speaker 2 (21:10):
Kind of good open minds, that type of thing.
Speaker 6 (21:13):
So it's it's been a very useful in a practical sense,
and I'm sure we'll continue to get you more.
Speaker 2 (21:19):
So, so I mean at the most, I shouldn't say
it the most. In one sense, it's doing a lot
of the menial tasks that it sounds like other people
don't want to do.
Speaker 10 (21:27):
Okay, So imagine for a company using AI. An example
would be the company I used to run. There was
a mortgage company. I was an IT and using eight AI,
customers could go in and put in their circumstances you know,
where they have veteran were they this, were they that?
(21:47):
And at the end of the test by AI, it
would say, oh, you qualify for the following possible mortgages.
There's a need.
Speaker 2 (21:56):
I'm not saying there's not a need.
Speaker 1 (21:57):
I'm just curious how I hear the job accounts and
my eyes glaze over.
Speaker 11 (22:04):
Hi Gary, I use chat GPT regularly for work. For example,
if you're working on an Excel programming need a formula,
you can type in the jat chat GPT what you're
trying to do, and if you want to change or edit,
you just respond to the chat GPT and it's like
you're having a conversation with it, unlike Google when you're
(22:27):
not able to do that. I hope that helps.
Speaker 4 (22:29):
Yeah, day Billy Eman here lived in Sandy. I'll list
you guys a long time Texas in California, moved to Colman, Alabama,
and I do water treatment and AI does incredible calculations
chat GPT specifically for dosages.
Speaker 5 (22:47):
Of chemicals, system volumes, for circulation rates. Chiller Tonnage. I
was hesitant to use it at first, double check the work.
It shows its work and it's right on.
Speaker 4 (22:59):
Thank you, love you.
Speaker 2 (23:00):
I mean to think about it. There's just every industry
that we've heard from, not including ours, where people use
it on a regular basis.
Speaker 12 (23:07):
Hy Gary and Shannon we're wondering who uses AI. I
love AI. I have dyslexia as a grown woman, and
this has been the greatest thing for me when writing
emails doing things for my company, and I put in
basically what I would like to say, and it fixes
It puts things in order, and it makes the emails
(23:29):
and everything flow so beautifully.
Speaker 11 (23:32):
So grateful for it.
Speaker 2 (23:33):
HEYI all right, and one more.
Speaker 13 (23:34):
Hey, Gary and Shannon absolutely adore your show. I actually
use chat GBT on an almost daily basis. I own
a consulting company and I use it for analyzing complex
data sets. I use it for translating text into training materials,
and also use it a lot to help refine communications
(23:56):
and presentations. It's definitely a great tool. Hope you have
a great day. Love you.
Speaker 1 (24:00):
Complex Data Systems PC load letter with the FSPC load letter,
we're starting to break out in hives with all this
work speed.
Speaker 2 (24:10):
Your printer is out of paper. I don't know.
Speaker 1 (24:14):
There was a lot of words there I could not grasp.
Speaker 2 (24:17):
Oh god. All right, you've been listening to the Gary
and Shannon Show. You can always hear us live on
KFI AM six forty nine am to one pm every
Monday through Friday, and anytime on demand on the iHeartRadio
ap
This is Gary and Shannon and you're listening to KFI
AM six forty, the Gary and Shannon Show on demand
on the iHeartRadio app. That's just said, that's not sad,
it's a data think about her and to celebrate her life.
Speaker 2 (00:14):
But I didn't think about it until I just looked
at the rundown. Oh.
Speaker 1 (00:19):
I mean, well, you haven't thought about twenty eight. I
haven't thought about the between twenty eight, right. It'd be
different if you saw it earlier and just forgot.
Speaker 2 (00:27):
We're finally doing a burial. If I told you that,
Oh good, when is that? That's gonna be next month?
That's nice? Hey, what's the rush?
Speaker 3 (00:36):
I know?
Speaker 2 (00:36):
And that seems to be the families. That's that's the
way we've always done it. Come on, we don't have
any religious obligations to bury people within twenty four hours,
so we wait, you know, fourteen months or in some
cases two years.
Speaker 1 (00:49):
Are why did you wait so long? It's not like
you're planning a party or anything.
Speaker 2 (00:55):
No, and it's just going to be it's just going
to be family. Yeah, So I don't know, I don't
know why. How do you make those decisions? I mean,
that's first of all we had to figure out where
because there's two family there's kind of two family cemetery things,
and one is mom's generally mom's side, the other one
is generally Dad's side.
Speaker 1 (01:14):
I can see if this was a you thing where
you're kind of like, eh, whatever, But your sisters seem.
Speaker 2 (01:18):
Kind of type A to me. Yeah, I mean there
is that.
Speaker 1 (01:22):
So that's kind of different to me that they would
wait so long. Yeah, well, but there's been a lot
of there's been a lot going on, a lot, there's
been a lot going on, so it's probably fine. It's
actually kind of nice that you can do it with
all the other crap to get done with the house
and the property and all that. Now that that's been
on that.
Speaker 2 (01:40):
True, Yeah, that is true. So anyway, not to start
on a negative note.
Speaker 1 (01:44):
Wow, well you want to do more dead stuff?
Speaker 2 (01:46):
We could. I found out that if if a bible
becomes damaged, the way that you dispose of it is
however you want to dispose of it.
Speaker 1 (01:58):
Why would you dispose of it just because it was damaged.
Speaker 2 (02:00):
Well, there was a time in the most recent past,
I'd say within the last week that my dog shredded
a bible.
Speaker 1 (02:09):
Really, so it wasn't just the yoga blocks.
Speaker 2 (02:12):
It was not just the yoga blocks. In fact, this
was pre yoga block. He went after and just destroyed
a Bible that was sitting on the couch. What do
you think that means? I do not know, but I
did ask for pastoral advice. I texted a friend of
mine who was a pastor, and I asked him what,
first of all, what I should do with dog? And well,
(02:33):
clearly the dog is not redeemed. Put the dog down.
And then I said, well, how do I dispose of
a Bible? Is there like is there.
Speaker 1 (02:42):
Is there like some sort of turn it in?
Speaker 4 (02:44):
Well?
Speaker 2 (02:45):
Ak, is it like an American flag? Like there's there's
procedures and policies and physicians. And he said, yeah, put
it in the recycle bin. Was a jump dum.
Speaker 1 (02:58):
It's a book. So it was sitting on the couch.
Speaker 2 (03:02):
Huh. That does that mean that you regularly read it?
My wife was reading it? Oh good, Yeah, that's good.
Speaker 1 (03:08):
At least there's one of you in that house that
is God fearing and.
Speaker 2 (03:13):
I have a Bible. Yeah, but you don't read it.
Speaker 1 (03:17):
In the other room behind behind the gate. Why is
the Bible behind the.
Speaker 2 (03:22):
Gate so that the dog doesn't chew it up. When's
the last time you picked up your Bible? You mean
the book or in my phone? I have an app
on my phone.
Speaker 5 (03:30):
Oh you do?
Speaker 2 (03:31):
Okay, I bring up right there, Matthew fifteen is what
it was. Oh okay. Some ps two Psalm two, Verse one,
Why do the nations rage and the people's plot in vain?
The kings of the earth set themselves and the rulers
take counsel together against the Lord and against his anointed.
Speaker 1 (03:53):
That's exactly what I was thinking of when I was
thinking about this Ai fight between well, China and everybody else.
This because the kings of the world of Ai, they're
doing what the Psalm says.
Speaker 2 (04:05):
Well, and they're doing a kind of behind curtains. I
don't want to say behind closed doors, because there's some
information that we know, but there's so little that we
know about the generation of deep seek Ai. Why I
mean this it is, it's not just chess.
Speaker 1 (04:24):
Here's China playing chess with our checkers. You know, Biden
going out and you know, banning companies here in the
United States from exporting AI chips to China was kind
of like, we're not going to give you our super powerful,
super expensive chips, and China's like f us, no f u,
We'll do it on our own for cheaper. We don't
(04:46):
need your chips. We look like fools in this, like
we thought that everyone needed our super expensive chips to
go and make their AI advances and that's not the case.
We look like the dumb dums on the global state
right now. And Trump is right that this should be
a wake up call.
Speaker 2 (05:05):
And then all it takes is a Chinese mcguiver to go,
oh well, I don't have the greatest, the fastest, the
biggest chips. Give me a ballpoint pen, ballpoint pen, some
duct tape, and then all.
Speaker 1 (05:17):
The trans transmission it used to be a transmission. Tranny
was a word for a transmission before I do any other,
meaning let's.
Speaker 2 (05:25):
Bring that back. So yeah, there's a lot that's that's
going on. Also, I have a question that I just
thought of, who is using this on a regular basis?
Who uses chat GPT?
Speaker 1 (05:38):
I think ki, I think it's gonna be everyone's using
it in the future.
Speaker 2 (05:43):
Right, Well, I mean because to me. Okay, so there's
a couple there's a couple of things. This morning I
realized Wall Street Journal now summarizes their articles. There's three
bullet points, and I think a lot of news organizations
do this now that I recognize it. Three bullet points
at the top of a long article. You can read
those three bullet points and get an idea of what
the article is. And there's a little thing in the
(06:04):
corner that says, what's this. I clicked on that this morning,
and it says, this is an AI generated summary of
the article that you're about to read. Well, no one's
going to read it if AI already just summed it
up in three bullet points. That's one thing. The other
thing is Google, when you search for something now automatically
gives you the AI version compendium of everything it's searched for.
(06:29):
Apple is trying to get you to update all of
your iOS and operating system stuff with its AI version.
It jammed down your throat.
Speaker 1 (06:39):
I was shocked, and not so shocked with the quickness
that I adopted Google's AI search results. When they first
started doing that, I said, what the F is this?
Screw this, I'm not paying attention to that. I would
scroll down, I would click on the various different things
that popped up to reach the summation that the AI
(07:00):
bullet points would have given me right off the bat.
So now, I mean it was probably a matter of
days where I was putting up a strong front of
I'm not using AI, I'm not using Google's AI, the
AI bullet points are feeding me, I'm not using that.
It was probably a matter of like eight to twelve
days between that and me gobbling it all up.
Speaker 2 (07:22):
Can't stop it because it's easy and we're lazy people, Okay,
but I mean, who's For example, this deepseak ai has
skyrocketed over the last three or four days to the
top of Apple's App Store downloads. Why what are people
just curious about it? Because when chat gpt came out,
(07:42):
people were like, make a poem about a monkey, hump
and a deer, and.
Speaker 1 (07:47):
Make a poem about penis, and people went with that
because it was a curiosity, right. I think a lot
of people use this for work related purposes.
Speaker 2 (07:57):
Oh yeah, school, I will say this. I did ask
chat gpt to write a radio play for our holiday
show and it was awful.
Speaker 1 (08:10):
Okay, so here's something that you should know. We don't
have real jobs, right, so we can't fathom how people
would use this in a practical and time saving work
smarter not harder manner.
Speaker 2 (08:23):
I'm hoping this is the I want to learn. I
want if you use AI on a regular basis, whether
it's Chat, GPT or deep seek or whatever it is,
if you use it on a regular basis, I'd like
to know how you can leave us a message on
the talkback feature on the iHeart app. Wall Street is
some you know, turning to some sense of normalcy after
(08:46):
what happened yesterday when, for example, Nvideo, one of the
big AI chip makers, lost more than five hundred billion
dollars in market value in one day, five hundred bit.
It had five hundred billion to lose, and it lost
five hundred billion dollars.
Speaker 1 (09:04):
A couple bits of feedback. Look at David Muir's muscles,
My goodness, is that necessary to put those out?
Speaker 2 (09:13):
It's a hot day in that helicopter. That's a G
I Joe commercial, right there, wasn't it? That's exactly what
it was. Okay.
Speaker 1 (09:22):
We've gotten a few bits of feedback with your dog
eating the Bible.
Speaker 2 (09:26):
One bit of feedback.
Speaker 1 (09:27):
Is Gary's dog is metal as f okay. The other
is that you should knock out five hail mary's that
you did not protect the Bible, that the dog was
doing dog things.
Speaker 2 (09:39):
But it's up to you to protect the Bible. So
protect the Bible.
Speaker 1 (09:42):
So you're good for five hail marys, which I think
is low. I mean I would go with, yeah, protect
the Bible. I don't do hail mary's. That's n what's
a Christian prayer? Just like our father? You do that one,
but you do the our Father right, everyone does, sire?
Speaker 2 (09:56):
Yeah, everybody kind of does that one. All right, how
about give me five ur our fathers will call it again.
That's not how it works. It is no, no, no,
you don't pay a toll to get across the bridge.
Speaker 1 (10:06):
Well no, you've already crossed the bridge and then burned
it down, so you now have to pay the damage
fee for the Bible.
Speaker 2 (10:15):
Yeah it is good morning, Hey, Ary.
Speaker 6 (10:20):
I know you, you know your upbringing was a little
unorthodoxed and you know, some weird stuff, But I think
it's all cool. But the fact that you guys aren't
going to bury your mom next to your dad, I
find that kind of strange. I think that's a no brainer.
Speaker 2 (10:35):
You know if I'm wrong, aren't you?
Speaker 6 (10:37):
But I can't see any other place. I think that's
the most appropriate thing to do.
Speaker 2 (10:42):
Feel very strongly about people. He doesn't know that. Well.
First of all, f you, you don't get to tell
me where they're gonna They get to decide where they're
going to be buried. And yes, they're going to be
buried together. That's such a weird.
Speaker 1 (10:57):
I'm pretty sure my husband has it somewhere in writing
to not be buried next time.
Speaker 2 (11:04):
Don't care what you do in life separated and give
me some f and pece around here. That's great. Good morning, Gary.
Speaker 7 (11:13):
Hey, did you ever get rid of that squatter in
your parents' house?
Speaker 1 (11:16):
Yes, yes, that was That was one of the things
you had to take care of before the burial.
Speaker 2 (11:21):
They would not have wanted us to have that around.
Speaker 1 (11:24):
Hi, Gary, I love you both.
Speaker 8 (11:26):
Hey. I am a nurse case manager for an insurance company.
I work in their workers' comp division, and we use
AI every day to kind of as a collaborative partner
for medical reports, looking at diagnostics and anticipating what type
of care a patient or an injured employee might need.
Speaker 2 (11:49):
Okay, talk to you so well, that's good. I mean
that makes sense. We're going to talk in a couple
of minutes here about what Deep seek Ai is, why
it caused such an earthquake on Wall Street yesterday. But
this was a comment from President Trump just a few
days ago.
Speaker 3 (12:03):
The release of Deepseek Ai from a Chinese company should
be a wake up call for our industries that we
need to be laser focused on competing to win. Because
we have the greatest scientists in the world. We always
have the ideas we're always first.
Speaker 2 (12:18):
We may always be first, But China did it cheaper.
Speaker 1 (12:22):
It's time for us to stop thinking that way, because
it's just not the case anymore.
Speaker 2 (12:27):
I don't think it's the case anymore.
Speaker 1 (12:28):
If this is real, if China jumps out and says
we can do it for cheaper, or we thought that
we held the key to the expensive computer chips that
make this all possible. I mean, as I mentioned, the
Biden administration band companies from sending these powerful AI chips
to China. So we must have thought that we were
(12:49):
the big kid on the block.
Speaker 2 (12:51):
We're not. If they're able to do this, you raise
a great question though, if this is real. Remember this
is a China needs product, and with that comes a
certain amount of mystery about its background. About its origins,
about how it was actually created, who funds it, and
what kind of information is it going to take when
(13:12):
you use it? Where does that information go? Yes, echoes
of TikTok.
Speaker 1 (13:16):
They have shown some of their math. They did publish
a technical paper last week about exactly how they were
able to do this PS. They're making it available for
other people to get in there and tweak it or
change it.
Speaker 2 (13:30):
And here's why. They want as much.
Speaker 1 (13:32):
Information about us or you or the smartest of the
smart that they can get. And if you get into
their system and you start tinkering around, guess what they're
tinkering with all of your stuff. It looks like most Americans,
more Americans, I should say, are working two jobs. Rising
caused a strong side market. Do you have a side gig?
A side market? Are you selling feet picks? What's going on?
(13:54):
There's no judgment here.
Speaker 2 (13:56):
Yeah, they would be total judgment. No, there wouldn't.
Speaker 1 (13:58):
Hey, if your feed are nice and have to send
feet picks, I mean I would have to pay people
to look at my feet.
Speaker 2 (14:04):
They're so gross. Be careful what you'd say. People are
into gross feet people lid for every pot. Wow, that's
really bad. I never thought about that. An eyeball for
every corn.
Speaker 1 (14:19):
Also, at the top of the hour, we're going to
tak an eyeball for every corn.
Speaker 2 (14:23):
Yeah, the corns on your feet.
Speaker 1 (14:24):
Oh oh, I don't have corns bunions, No, but look
co fungus. Look, oh god, they're not great.
Speaker 2 (14:36):
Ah here, wipe off the You're gonna throw a bare
foot up there. Also at the top of the hour,
the creek fire. If you remember that from twenty seventeen,
so cal Edison said, hey, it wasn't our power lines.
The FEDS think so cal Edison may be lying about that.
(14:58):
Of course you think of that in the context of
the eating fire, and now people pointing their fingers towards
the utility for that one as well. So we'll talk
about that coming up.
Speaker 7 (15:05):
A game changing move that does not come from Open Ai,
Google or Meta. There is a new model that has
all of the valley buzzing, but.
Speaker 2 (15:14):
From a Chinese lab called deep Sea.
Speaker 9 (15:17):
It's opened a lot of eyes of like what is
actually happening in ai in China.
Speaker 1 (15:23):
Deep Seek is a Chinese AI startup. It was founded
in twenty twenty three by Langwin Fang. He's the co
founder of a hedge fund high flyer. High Flyer uses
AI algorithms to predict market swings, so you can see
why he's got stock in the game, right. If you're
going to use AI to protict market swings and make
an s ton of money, not for yourself alone, but
(15:46):
for everybody else, you want your AI to be tip top.
Speaker 2 (15:50):
You want it to be accessible and cheap. So that's
what he set out to do.
Speaker 1 (15:55):
They claim that it's AI software is competitive with many
main made by US tech giants like open Ai and
Google at a fraction of the cost. Last week, Deep
Seek did publish the technical infrastructure, which talks about the
AI model behind the app. It claims cost savings from
(16:18):
using far fewer and less advanced computer chips than advanced
AI projects usually do, like the ones we do. It's
punitively expensive here in the United States because we've thought
all along that we need to use these advanced computer
chips that are more expensive.
Speaker 2 (16:34):
Turns out you don't. That is the biggest issue.
Speaker 1 (16:38):
That's the biggest gift though to us as well. If
we can harness that and create our own AI infrastructure cheaper,
it's a win for everybody, isn't it.
Speaker 2 (16:49):
Chaitan Puta Gunta is a general partner at Benchmark, and
I heard him on CNN explaining, we still don't know
a lot. I mean, as much as we do know
about deep Seek, there's plenty other questions.
Speaker 9 (17:00):
So did they actually assemble this talent, how did they
assemble all the hardware? How did they assemble the data
to do all this? We don't know, and it's never
been publicized. And hopefully we can learn that.
Speaker 2 (17:10):
Yeah, hopefully we can learn that. So here's a terrifying thing.
Alexander Wong is a founder of Scale Ai, and he
actually that company is helps AI systems learn. It kind
of puts together the information that AI then dives into
to learn from. And this is a terrifying aspect of
(17:34):
There are two parts of his answer here that are terrifying.
Speaker 7 (17:36):
Today we released Humanity's Last Exam, which is a new
evaluation or benchmark of AI models that we produced by
getting math, physics, biology, chemistry professors to provide the hardest
questions they could possibly imagine. Deep Seek, which is the
leading Chinese AI lab. Their model is actually the top
(17:58):
performing were roughly on par with the best American males.
Speaker 2 (18:02):
And again I mean that point of it's using it
with just a tiny percentage of the money that we've
put into AI with chips that we I mean, the
lower quality, slower chips that we have been thinking, Will
they never be able to put together an AI system
if they use those low quality chips? Can I can?
(18:22):
I just want to point something out here, he referred to.
He referred to something called Humanity's Last Exam. This is
a this is a test that they put forth for
these different AI models and the Center for AI Safety.
(18:43):
It's a test that they say as a groundbreaking AR
benchmark designed to test the limits of AI knowledge at
the frontiers of human expertise. And as he mentioned, they
just ask some of the smartest people in the world
come up with the absolute hardest questions that you can,
some of which may not act actually have answers that
we know, and see if AI can do anything with it.
(19:03):
I'm going to give you one example of the kind
of question that they ask AI to come up with
an answer for question one. Hummingbirds within a podiformis uniquely
have a bilaterally paired oval bone a sesamoid embedded in
the quadrilateral portion of the expanded curetiate upeneurosis of insertion
(19:28):
of M depressor kaudia. How many paired tendons are supported
by this sesamoid bone answer with a number twenty three.
I don't know if that's true. I don't know if
that I.
Speaker 1 (19:43):
Do deep Seek in terms of if it's safe, by
the way, that's very convoluted that question, like yay for
their AI if they can. They also ask about you
should know the answer.
Speaker 2 (19:53):
To biblical translations from a Hebrew in Greek mythology, who
was Jason's maternals, Jason's maternal great grand.
Speaker 1 (20:02):
Thoughts, So it's got a wide breadth of knowledge there.
Deep Seek, by the way, in terms of safety, can
collect and store a massive amount of personal information, including
any conversations you have, technical information about things like your device,
your Internet connection. This data can be accessed by the
Chinese government.
Speaker 2 (20:22):
Okay.
Speaker 1 (20:24):
The app also appears to be censoring certain information like
I don't know Tienneman Square and the protests there. Yeah,
right in the line of what China does now. Deep
Seek has released a version of its AI model for
others to use and modify. Outside developers can make it
more secure. Both more secure and less restricted by running
(20:45):
it locally on a powerful enough device. But yeah, the
fact that they've made it freely available for others to
download and modify means that they just want more people's
information in its clutches.
Speaker 6 (20:58):
Actually just USEPT in my car this morning.
Speaker 10 (21:02):
I'm on the way to go do some sales calls, and.
Speaker 6 (21:04):
I bounced some ideas off of chat GPT as far
as the message I want to get across my clients.
Speaker 2 (21:10):
Kind of good open minds, that type of thing.
Speaker 6 (21:13):
So it's it's been a very useful in a practical sense,
and I'm sure we'll continue to get you more.
Speaker 2 (21:19):
So, so I mean at the most, I shouldn't say
it the most. In one sense, it's doing a lot
of the menial tasks that it sounds like other people
don't want to do.
Speaker 10 (21:27):
Okay, So imagine for a company using AI. An example
would be the company I used to run. There was
a mortgage company. I was an IT and using eight AI,
customers could go in and put in their circumstances you know,
where they have veteran were they this, were they that?
(21:47):
And at the end of the test by AI, it
would say, oh, you qualify for the following possible mortgages.
There's a need.
Speaker 2 (21:56):
I'm not saying there's not a need.
Speaker 1 (21:57):
I'm just curious how I hear the job accounts and
my eyes glaze over.
Speaker 11 (22:04):
Hi Gary, I use chat GPT regularly for work. For example,
if you're working on an Excel programming need a formula,
you can type in the jat chat GPT what you're
trying to do, and if you want to change or edit,
you just respond to the chat GPT and it's like
you're having a conversation with it, unlike Google when you're
(22:27):
not able to do that. I hope that helps.
Speaker 4 (22:29):
Yeah, day Billy Eman here lived in Sandy. I'll list
you guys a long time Texas in California, moved to Colman, Alabama,
and I do water treatment and AI does incredible calculations
chat GPT specifically for dosages.
Speaker 5 (22:47):
Of chemicals, system volumes, for circulation rates. Chiller Tonnage. I
was hesitant to use it at first, double check the work.
It shows its work and it's right on.
Speaker 4 (22:59):
Thank you, love you.
Speaker 2 (23:00):
I mean to think about it. There's just every industry
that we've heard from, not including ours, where people use
it on a regular basis.
Speaker 12 (23:07):
Hy Gary and Shannon we're wondering who uses AI. I
love AI. I have dyslexia as a grown woman, and
this has been the greatest thing for me when writing
emails doing things for my company, and I put in
basically what I would like to say, and it fixes
It puts things in order, and it makes the emails
(23:29):
and everything flow so beautifully.
Speaker 11 (23:32):
So grateful for it.
Speaker 2 (23:33):
HEYI all right, and one more.
Speaker 13 (23:34):
Hey, Gary and Shannon absolutely adore your show. I actually
use chat GBT on an almost daily basis. I own
a consulting company and I use it for analyzing complex
data sets. I use it for translating text into training materials,
and also use it a lot to help refine communications
(23:56):
and presentations. It's definitely a great tool. Hope you have
a great day. Love you.
Speaker 1 (24:00):
Complex Data Systems PC load letter with the FSPC load letter,
we're starting to break out in hives with all this
work speed.
Speaker 2 (24:10):
Your printer is out of paper. I don't know.
Speaker 1 (24:14):
There was a lot of words there I could not grasp.
Speaker 2 (24:17):
Oh god. All right, you've been listening to the Gary
and Shannon Show. You can always hear us live on
KFI AM six forty nine am to one pm every
Monday through Friday, and anytime on demand on the iHeartRadio
ap
Gary and Shannon News
Popular Podcasts
40s and Free Agents: NFL Draft Season
Daniel Jeremiah of Move the Sticks and Gregg Rosenthal of NFL Daily join forces to break down every team's needs this offseason.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.