What Nvidia JUST Said [GTC Keynote Summary] YIKES Tesla!

well the GTC keynote from Nvidia is over

and in this video I'm going to give you

a complete summary as quickly as

possible in the over 2hour keynote

presentation of the new chips that were

introduced from Nvidia as expected

project code name Blackwell from a

famous game theorist who's no longer

with us died about 13 14 years ago at

this point uh they nicknamed or code

named this chip after uh this game

theorist and the idea was this is is the

next breakthrough for NVIDIA it's a chip

that's faster it's more efficient we'll

do some car comparisons in a moment but

basically you get a whole new stack

think server system assembled by a

company like either Dell or super micro

computer uh new uh Graphics GPU chips

throw in the uh gray CPUs that Nvidia

uses new chipset for linking all of the

gpus together uh new security modules

for linking these together and crypting

all of the computation throughout all of

these gpus together it's essentially

going from the iPhone moment of the h100

to the iPhone 2 which if I remember

correctly was the 3G but anyway so it's

sort of like your next year's iteration

I want to before I actually get into

some detailed summary just hit numbers

here for a moment how does this actually

change what we saw today the trajectory

for revenues at Nvidia what doesn't this

is just the next phase the question is

how long can Nvidia continue to sell

these chips and grow that's the biggest

question and that's not an answer you

can really get from this presentation in

a number format but you could see how

vast Nvidia is expanding its user

applications to try to make sure that

they don't stop growing anytime soon so

you're going to see that as I go through

the summary first in the first 5 minutes

we had self-driving cars mentioned two

times and later in the video we act or

presentation we actually learned that

byd is going to partner with Thor uh

which is the Nvidia uh sort of Omniverse

and full self-driving software uh that

Nvidia puts together and byd imp uh

plans to implement Thor in their

vehicles in the future the first vehicle

to implement Nvidia self-drive

self-driving features will probably be

the Mercedes glr at the end of this year

or early 25 but it looks like Tesla's

biggest competition might not actually

be another car manufacturer themselves

it could actually be the world's next

biggest or potentially the biggest auto

manufacturer combined with Nvidia so

while FSD 12.3 is pretty awesome I have

to say and we haven't seen any video or

footage of Nvidia does seem like Nvidia

is gearing up to be one of the biggest

competitors now you have to think about

this I just I'm purposefully starting

with this because I want you to think

about this think about the compute Tesla

uses just to train FSD now imagine the

compute generated by All Automotive

companies moving to AI GPU systems with

maybe invidious core package that they

can build off of and now everybody has

FSD you basically have democratized FSD

and you're writing and using Nvidia

chips later in the presentation they

talked about how you you could go to ai.

nvidia.com and you could basically

download different tools I summarized

all of this at ec.com by the way

including the link to ai. nvidia.com and

you could use different tools there one

of the things that I was playing around

with was I jumped into grabbed I grabbed

my cookies I I did I did it but anyway I

was messing around with their weather

tool over here uh and there are plenty

of different use cases and Integrations

and products you can use here and you

can actually play with some of these AI

products and see the demonstrations that

they can produce for you for certain

probably mostly at this point

pre-programmed events but they kind of

what they're trying to do is show you

look we're going to give you the AI tool

and y'all build off of it the idea is

can we give you a GPU uh GPT rather

that's based on your company's

infrastructure your company's systems

and procedures your stock brokerages

systems and procedures your doctor's

offices systems and procedures and then

can you take our little base thing turn

it into your own own stuff and then

utilize that to be more efficient all

while of course running it all on GPU

gpus put together by Nvidia that's sort

of the idea here and so I think when you

when you realize just the car Revolution

that's

coming excuse me which Nvidia is honing

in on is just one of the angles of

attack they have consider as we saw in

the presentation protein folding biology

genomic sequencing it's a whole another

world

Aviation consider uh construction

manufacturing all of these very

important forgive me for a second

SEC a okay hard to do a summary when

when you have a cough but anyway uh a

lot of shoutouts here in human robotics

as well that is humanoid robotics we got

multiple shoutouts and demonstrations of

Amazon Amazon was probably thrown up on

that screen 10 different times during

the presentation apptronics

robotic startup in uh Texas they built

humanoid robots they um I've actually

visited them in person really cool

seeing the robots in person spoken with

the CEO in person fantastic company they

got two shoutouts in this video multiple

different robotics companies got shout

outs in this video uh and especially

highlighting the self-driving

competition that's coming now keep in

mind we didn't get real demonstration

yet demonstrations yet right Tesla's

already demonstrating that's a big deal

for Tesla but what we're finding is

Tesla's biggest competition might not

actually be Tesla itself and a lot of

folks are going dang didn't even

consider Nvidia coming multiple

Partnerships were announced including

Partnerships between Nvidia and tsmc so

tsmc can use Nvidia AI in manufacturing

chips but intel was not mentioned Taylor

Swift got a shout out and the actual uh

Blackwell b00 GPU stack will be built on

the arm core processors it's basically a

giant GPU and there were some cool

Technologies they mentioned such as

using two blocks of semiconductors that

they were able to layer together to dyes

so to speak and they gave a lot of

technical examples I'm going to save you

most of the technical jargon because if

you really wanted the technical jargon

you'd go to somebody that's really

technically Advanced I'm going to be a

little bit more high level here uh

mostly because we're looking at this

from a financial analyst uh point of

view we've got this idea that we're

actually progressing with technology

faster than ever before ever before that

previously we were growing at uh

basically 5x sorry 10x every 5 years

100x every 10 years well we just went

eight years from the moment open AI got

their first uh I think it was the dgx uh

GPU machine essentially and from there

we've gone about a,x in just eight years

Jensen argues

this uh excuse me all right yeah liquid

cooled machines an example of how uh

this could be productive is that the

b00 would take about 90 days to train a

GPT but would just use 1/4 of the gpus

and so you could train the same model

with 1/4 of the gpus I think that's a

really like relatable stat so I put a

highlighter on that 1/4 the gpus needed

to do the same progress now of course

you don't only want to do the same you

want to do better right so you could do

that better work more

efficiently uh chat Bots tokenomics we

talked about llms the AI Factory Amazon

Oracle Google Microsoft all partnering

uh obviously the B100 will offset h100

sales when it's available we didn't

really get a release date a lot of talk

about Omniverse we've already known

about the Omniverse for a very long time

with Nvidia they've been pitching the

Omniverse for like 5 years now I feel

like excuse me it's basically a way of

having uh a virtual environment of a

factory so you can more appropriately

lay out a factory to be the most

ergonomic and efficient for workers uh

that is a lot easier than obviously

building a factory and then trying to

re-engineer it I hate to say it but I'm

not sure that Elon uses Omniverse

because one of the reasons Elon doesn't

like the phrase copy and paste factories

is because he says he learns how to do

the factories better every single time

they make them well the idea of the

Omniverse is that rather than building a

new Factory and then going ah we could

do this better this better this better

and then building another Factory that's

better you could do that virtually and

basically virtually iterate how you want

to build your factory and in the long

term robots and humans and whatever can

all work in a factory together

seamlessly being notified of emergencies

or issues or the best processes or

efficiencies for doing things they did

mention that you're going to be able to

stream the Omniverse from the Apple Pro

Vision Vision Pro dude I always say that

backwards I'm sorry I piss a lot of

people off when I say that I actually

have them right

here uh but anyway uh I I personally

really like them I know a lot of people

poo poo on them but I think they're

really cool and they're really great so

I'm a big fan of these uh for me this is

actually really critical when it comes

to distraction free work uh and I get

really burnt out as an example of like

sitting in this studio here I get really

burnt out sitting here and so I can

actually stay sitting in here because it

is a soundproof room put this on and

then feel like I'm somewhere else and

really just hone in like if I got to

finish something that I know I'm going

to have to write out it's going to take

me three hours to write out I could go

on this and and do that so I think this

is actually really cool that they're

integrating Omniverse in this I'm not

sure if they're going to like Stream It

Through Safari like you know type in the

IP address sort of thing which that kind

of seems lame and like it would be

clunky I'm sure it'll get better in the

future uh and and so something to think

about here is a lot of these

Technologies aren't going to be perfect

today right A lot of people I think are

going to see this high level vision and

go my gosh it's all going to be perfect

today no just like GPT isn't perfect

it's great but you know there are a lot

of times where like ah I know GPT is not

going to do well at that you know it's

great for some things but it's not great

for everything like Jensen himself said

gpts basically just imitate you they try

to understand your context and then they

imitate

humans that can have problems right that

can come with problems

so in the long term this is the future

though humanoid robotics uh you know

basically making vaccines or uh genomic

solving genetic diseases genomic

sequencing all of these things will be

done with AI in the long term there's no

doubt about that I personally I I think

we're going to have a buy the rumor sell

the news on the stock but for the very

long term would I be bullish Nvidia for

the very long term oh you you can't not

be bullish Nvidia you just cannot be

with 7 7% gross margins that'll settle

down to 75% over the next two quarters

this is phenomenal and uh I understand

the stock has run up a lot and and I

have long exposure to the stock but uh

and I do think it's going to sell down

in the near term uh along with probably

some of the other overblown chip sectors

spe specifically the super micro

computers the arm lock up uh and you

know these are all these are all

short-term considerations so if you're

trading yeah by the rumor s news if

you're really really long is this

exciting and and full of enthusiasm

absolutely the potential is Limitless

for example I really enjoyed their uh

demonstration when they talked about

predicting weather at accuracy right now

they say we can take predict weather

within a 25 km band of accuracy but

that's actually pretty dang wide I

always think about hurricanes because I

grew up in South Florida where we had a

lot of hurricanes 25 km okay so what is

that like 15ish miles or whatever well

that that's a big spread and so what I'd

like

uh is more accuracy and what does Nvidia

say they're getting to 2 km accuracy

that's great that's fantastic and maybe

we could predict earthquakes or

hurricanes or you know they call them

typhoons in different parts of the world

better they're just starting with

Typhoon typhoons over uh by Taiwan right

now they got to protect

tsmc I got to protect their chips but uh

uh but you know this will all expand in

the long term now I feel like I'm losing

my voice

too okay uh then they've got Nims Nvidia

inference microservices basically

pre-trained packages we kind of already

touched on that uh chat Bots robotics

the physical world sensor processing uh

air traffic control uh I think uh Autos

autonomy I mean the use cases here are

quite frankly endless uh and it's really

really enjoyable to sort of think about

these things again the Mercedes glr

partnering in early 2025 no

demonstration of the ability of this

vehicle yet but it's coming and uh

Tesla's remaining Mo feels like it's

quite frankly manufacturing Vehicles

hopefully we'll see what their margins

end up like uh hopefully manufacturing

robots and uh an FSD but FSD has got a

competitor coming though I think Tesla's

way ahead for now humanoid robotics

Nvidia Project named Groot General

robotics 003 is what that stands for so

that's gr00 T then at the end they

showed off two Wall-E style robots those

are probably remote controlled like I

don't actually think they're interactive

or autonomous by any means certainly not

during a live event there's no way so

but they were still cool seeing the

little Wall-E style robots come out uh

and so then they wrapped it up with a

little spaceship and micro kind of

flying around their animations were

great they did a fantastic job showing

some generative AI but uh look the

reality is can you really bet against AI

in the long run no of course not can you

bet on every single software company no

you probably are going to see like 90%

of software companies in AI fail you'll

probably see 70% of Robotics companies

fail and 70% of auton like you know uh

uh car companies fail there's almost no

doubt about

that but one thing that will be

consistent is that pix and shovel style

investment who's got the best chips to

actually operate AI on obviously uh you

know today's an Nvidia day uh AMD is

watching this very closely to make sure

not only can they catch up but they can

exceed Nvidia so obviously expect

something very similar from Lisa Sue

over at uh AMD and uh and that's my take

now again bottom line let's do a quick

little stock look is the stock going to

go up or down well you know the Stock's

already done exceptionally well it's

certainly done better than I thought it

would as quickly as it did uh I did

think it was going to go to $600 or $700

I thought this was realistic I thought

we were going to stop here and get

rejected I was wrong we got rejected at

the next line at the uh

969 this by the way is Weeble on Windows

if you want it go to metkevin me t k v i

n.com Weeble uh you'll get like 20 free

stocks fractional shares or whatever

they do it's worth up to $3,000 if you

sign up with that affiliate link so go

to that link before you go to

Weeble but anyway uh my take is um you

my really the same as what we had over

here is that I think this is a great

company to invest in the long term but

realistically I thought that the the

growth potential here is going to become

a little bit more limited in the near

term just because you uh you've already

priced in so much enthusiasm I don't

think this event really moves the needle

I was actually a little bit more

enthusiastic about potentially Apple uh

being a play I wanted to buy the dip on

Apple over here and I know it's moved up

a little bit today 64 bips it's been

moving up a little bit but I'm actually

really disappointed that apple was

considering using Google uh and Gemini

for their artificial intelligence I was

looking for something a lot more

proprietary here so I'm a little Tor I'm

going to have to do some more analysis

on Apple part of me just feels like we

need to wait for a little bit of a

correction on the cues thanks to JP's

rugging mostly because I've predicted

this Nike Swoosh style recovery but I've

always called it a volatile Nike Swoosh

there's been no volatility and so that

does make me a little nervous that

things have got a little euphoric here

since about November and maybe a little

bit so much so but what do I think is a

better investment in the long term

starting right now well I don't think

you can bet against the chip sector

obviously not personalized Financial

advice if there's a little correction

it'll get bought if Nvidia drops to 700

or whatever it is it'll get bought up

you know 730 I think is where the level

was yeah 731 it'll get bought and that

might be a great entry point for the

long term I'm not convinced that entry

point is here I personally would touch

super micro computer because there'll be

a lot of companies doing this I don't

think they have the IP I think Nvidia

has the IP uh

dell is very much in that super micro

computer boat for me Dell by the way I

did not expect that they would blow up

like this when they got their shout out

I mentioned them in my Nvidia review

video that they got a shout out from

Nvidia which they got today as well and

their stock blew up and so they got

another shout out today Stock's up 2%

big deal it's already kind of trending

down why because only 2% of this

business doubled so great 4% of this

business is coming from AI but the rest

of the business isn't doing too hot so I

have some problems with uh dell I

actually think in i' I'd be more

considering looking at if I get a sell

off on AMD I'm down to have a nice

barbell between like an AMD Nvidia uh I

like TSM uh and asml don't don't count

out asml making your Advanced

lithography machines don't count these

guys out they make the chip they make

the machines that make the machines

basically or make the chips that they

make the machines that make the chips

put it that way the uh

euv uh uh lithography machines they're

essentially the only company in the

world they've got nearly a monopoly on

this obviously interest rate sensitives

are a little little risky in this

environment but uh look long-term

bullish shortterm I don't even know that

I could really be more bearish than just

saying by the rumors H the news because

I don't think there's a bearish

near-term catalyst Beyond maybe you'll

get some sell down after this event's

over big deal but I don't see a

near-term bad Catalyst for chips I do

see a near-term bad Catalyst for

interest rate sensitives uh and I think

interest rate sensitives are going to

hurt for way longer than I could have

possibly anticipated because the economy

is just doing too damn well and not only

is the economy doing too damn well but

uh the Federal Reserve is going to be in

no rush to cut interest rates anytime

soon so that's um that's my take

hopefully this was useful for you if you

like my perspectives look I'm not always

right but I do my best to provide you

quality information every single day and

I extend that quality information my

course member live streams every day

where we brainstorm with fundamental and

technical analysis every single day and

real estate analysis join me there join

the stocks and psychology money group

it's linked down below all of the Le uh

courses get access to the course member

live stream and the stocks and sight

group gets byell alerts as well so

consider joining that link down below we

do have a price increase coming for the

courses as well as a price increase

coming for the meet Kevin event the

milli Symposium which is June 21st to

23rd uh and then last but not least if

you have not yet considered investing in

house hack and you're an accredit

investor and you want to diversify a

little bit away from some of the crazy

gains of chips take a little bit of

chips or a few chips off the table so to

speak consider investing in house Haack

at house.com

12024 so you could see our PPM for

investing house thank you so much for

being here we'll see you all in the next

one goodbye and good luck adver these

things that you told us here I feel like

nobody else knows about this we'll we'll

try a little advertising and see how it

Go congratulations man you have done so

much people love you people look up to

you Kevin PA there financial analyst and

YouTuber meet Kevin always great to get

your

take even though I'm a licensed

financial adviser licensed real estate

broker and becoming a stock broker this

video is not personalized advice for you

it is not tax legal or otherwise

personalized advice tailor to you this

video provides generalize perspective

information and commentary any third

party content I show shall not be deemed

endorsed by me this video is not and

shall never be deemed reasonably

sufficient information for the purposes

of evaluating a security or investment

decision any links or promoted products

are either paid affiliations or products

or Services we may benefit from I also

personally operate an actively managed

ETF I may personally hold or otherwise

hold long or short positions in various

Securities potentially including those

mentioned in this video however I have

no relationship to any issuer other than

house act nor am I presently acting as a

market maker make sure if you're

considering investing in house Haack to

always read the PPM at house.com