November 21, 2024 • 44 mins
Bloomberg's Caroline Hyde discusses the latest sentiment around Nvidia as investors cool on the AI darling despite posting an earnings beat. And, the new Trump administration is said to be mulling a first-ever crypto policy role. Plus, Palo Alto Networks sees easier "fixes" ahead in cybersecurity as the world moves from hardware to software.
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 (00:01):
From the heart where Innovation, money and power Collie in
Silicon Valley, NBN.
Speaker 2 (00:07):
This is Bloomberg Technology with Caroline Hyde and Ed lud.
Speaker 3 (00:10):
Love live from New York.
Speaker 4 (00:25):
This is Bloomberg Technology coming up in video.
Speaker 3 (00:28):
Says it's new.
Speaker 4 (00:28):
Chipper on track, but the rush to get the product
lineup out the door is.
Speaker 3 (00:32):
Proving more costly than expected.
Speaker 4 (00:34):
Plus, Bitcoin rises to new heights as the Trump transition
team weighs the creation of a dedicated white house cryptopost
and the CEO of Palo Alto Networks joins on the
heels of its earnings.
Speaker 3 (00:45):
We'll discuss how AI can.
Speaker 4 (00:47):
Help stop the threats of the future, but first we
get straight to the world of in video and indeed
the points trag that it's having on the Nasdaq is
somewhat We're off by eighteen percent, but it's been a
volatile day of trade as people.
Speaker 3 (00:59):
Try to pass.
Speaker 4 (01:00):
Ultimately, how strong future growth still looks under in video
and indeed the supply side headaches that much has been discussed.
Let's get straight to within King and look, we'd all
been fretting about Blackwell the product lineup, how easily it
is to get into customers. Hans did Jensen Wong manage
to calm investors' nerves?
Speaker 5 (01:19):
I mean, obviously to an extent. We haven't seen the
massive fall off that we've seen manifest previously with concerns.
He talked about, Yeah, we're getting it to market. We're
actually going to be probably getting more to market than
we had thought we would. At the same time, it's
going to cost us.
Speaker 4 (01:36):
Yeah, let's just weigh in on the profit margins, because
you do such a great job at giving the context here. Ultimately,
this is a company that gave us ninety four percent
increase in revenue in the third courting guided to another
seventy percent increase, give or take plus or minus two percent.
But the profitability is what just coming down ever so
slightly to what seventy three to seventy four percent margin?
Speaker 5 (01:57):
That's right, I mean, I think that's the right way
to look at it. They've set such high expectations, right,
they were at seventy five percent, which had narrowed slightly.
But seventy five percent is like software mhudgeons, Right, this
is a semiconductor company, right, you know, AMD's twenty points
south of that. You know, Intel isn't even in the
same zip code right now. So yes, they're losing a
(02:18):
few points because they're spending more in engineering to get
those chips out, get those systems in place quicker.
Speaker 2 (02:24):
But really this is a very very high bar.
Speaker 4 (02:27):
And then full production comes what the back end of
their fiscal twenty twenty five in. But I suppose they
still have to address the issue that. Sure they're getting
them out the door, but it's to the same kind
of companies. How are they managing to diversify their demand base?
Speaker 5 (02:43):
Well, I mean that was one of the things that
we saw in the release. They're not right. The amount
of revenue coming from those big companies, the Microsoft's, the
Amazons has actually gone up from forty five percent to
fifty percent, So like half of the most important divisi
that they have is dependent upon just a few companies.
Investors are definitely concerned about that. Why they're concerned about
(03:05):
that because these companies are making their own chips, trying
in effect to replace in video and everybody else that
supplies them.
Speaker 4 (03:13):
Is there any real competition here, because we've heard it
time and time again that as much as AMD tries,
as much as Intel tries, really in videos, the any
game in time for here.
Speaker 5 (03:23):
Yeah, I mean that that definitely appears to be the case.
I mean that any questions about and demand, any questions
about losing market share were just irrelevant, right. I mean
they talked about incredible demand, demand that they cannot meet
right now. So you know that part of the quote,
the story is really in the future, if at all, Ian.
Speaker 4 (03:45):
King very much in the present when it comes to
the market move, We really appreciate it.
Speaker 3 (03:50):
Thank you.
Speaker 4 (03:51):
Let's get you more contact and what it means to
the rest of the market. Epshkere is with US Senior
and a list at Swiss Quote and Ibet. We turned
to you for the market implications because we were anticipating
as much as a three hundred billion dollar market cap
swing up to eight percent move higher or lower. Actually,
a lack of holatility must be a slight sign of relief.
Speaker 6 (04:14):
Well, actually it is because there has been a few
red flags in yesterday's report. As A was just talking
about the pressure on the productivity levels due to the
manufacturing challenges for the black Bell chip is one of them.
The fact that the big technology companies make up to
fifty percent of Nvidia's revenue is another sticky point. Yet
(04:34):
the results have been strong and despite these small red
flags that we got yesterday from the report, well, the aftermass,
the immediate selloff has been only twenty half percent, and
this morning we don't see that the Nvidia shares are
being sold at the level that we would expect them
to with such a disappointment, or if with the small
(04:55):
disappointment from investors, we would think that any miss that.
Speaker 7 (04:59):
Would be a excel off.
Speaker 6 (05:00):
Yet we just see that the market is quite resilient.
Speaker 4 (05:05):
We've got a bit too used to him managing to
smash expectations.
Speaker 3 (05:09):
That he had set and that of the market.
Speaker 4 (05:11):
And look, you only beat them slightly when it comes
to their own internal guidance IPEC. The rest of the market, though,
had been anticipating, well, maybe a questioning around valuation. What
do you make of the fact that we're still up
let's call it two hundred percent over the course of
this year. It's an extraordinary run up for one particular company.
Speaker 3 (05:31):
Can it be sustained?
Speaker 6 (05:33):
Well, it is absolutely an extraordinary run up, and now
we are questioning the valuation levels, especially if.
Speaker 7 (05:40):
The demand started to solve.
Speaker 6 (05:42):
This is not the case right now, but at some
point we expect that the demand from the big technology
companies will start swing, and because this is a big
chunk of the company's revenue, that's going to be a headache.
Speaker 7 (05:53):
There's another thing.
Speaker 6 (05:55):
Right now, the competition is not a problem because Nvidia
has got the most expensive but the most premium chips.
They've got the best performance, and the big technology companies
are looking for those. But if the demand was to
slow for the big technology companies, then Nvidia will have
to find other customers. And these other customers may not
be looking for the premium solutions. They will be just
(06:16):
looking for the solutions to manage their day to day
businesses and increase their productivity and increase their costs. But
they would also up for maybe more cost efficient and
well slightly more affordable chip solutions than the Nvida offers.
And I think that that's going to be a major issue.
Speaker 4 (06:35):
We're going to be talking exactly that later in the programming.
For now, when we think about the context of in video,
about twenty percent of all the run up in the
S and P five hundred is thanks to this one
particular name this year, about twenty five percent of all
earnings per share increase. How key man risk is in
video and how does one protect themselves against that.
Speaker 6 (06:56):
Well, obviously, the AI rally and especially has been one
of the major pillars of the market rally in the
US that we have seen over the past two years.
So anything any lack of epissites for Nvidia from now
on is going to have a negative impact on the
market sentiment. Now, we have seen that AI rarely broaden
(07:18):
toward the other sectors, other non technology sectors. But it's
worth noting that this company has added so much upside
pressure to the SMP five hundred, So anything less than
any kind of loss of momentum here will obviously be
well quite negative for the SMP five hundred, especially knowing
that the SMP five hundred today is struggling near its
(07:41):
all time high levels.
Speaker 4 (07:43):
So when you're sat ultimately in Europe and you're thinking
about a global perspective, it should people be broadening out
when it comes to just magnificent seven, let alone just
in video litlone just US bets on technology.
Speaker 6 (07:57):
Well, when it comes to technology, I really believe that
the US is still in a very dominant position, So
you wouldn't be going to to Europe for technology solutions.
There's one place that we actually like, and that's Japan,
especially given that the government there is going to give
some more support to their technology industry.
Speaker 7 (08:15):
So in terms of geographical.
Speaker 6 (08:17):
Diversification, Japan could be a good solution. There is China
as well that we have been looking at, but we
are still not positive for China because well, the chipwoar
has weighed greatly on sentiment and on the progress of
the technology advancement there, So we're not really looking at
China right now. So one place that investors could be
(08:38):
looking at would be Japan, but other than that, the
US is still in a very dominant position. So if
you want to invest in tech, well you would just
the first place you would like to go is the
US because this is where you have massive upside potential.
Speaker 4 (08:50):
So back the clients that are calling you, are they
saying they want to add to Invidia, for example, as
a name at these particular levels or do they just
keep their exposure where it is well.
Speaker 6 (09:01):
To be perfectly honest with you, I don't have clients
who would like to sell Nvidia, but at the current
levels people are more skeptical about entering into.
Speaker 7 (09:09):
The NVDA shares.
Speaker 6 (09:10):
What I mostly here is what level would be a
good entry level if we start seeing a price pullback
from the actual level, So people will still be buying
MVDA shares from this point on, but they're still at
disvaluations at this point in time, they will be looking
for good entry levels.
Speaker 4 (09:29):
Is there other names, smaller chip rivals that you've thought
would be a better entry point.
Speaker 6 (09:37):
Well, I think that AMD is at a good position
right now. Obviously, they do not offer the same potential
and d do not offer the same margins than Nvidia does,
but it's share price have come down quite significantly to
offer an interesting entry point at the current levels. It's
also to be said that AMD is about to launch
(09:58):
two powerful chips, one for the end of this year
and one for next year. And what's interesting with AMD
is they may not be as powerful as premium as
Nvidia chips, but they're clearly going to be more cost
efficient and there are many many sectors and companies out
there that will be looking for more cost efficient and
more affordable solutions, and AMD could actually help fulfilling that gap.
Speaker 4 (10:22):
MD currently trading at forty one times future earnings compared
to about a fifty times for a video effect Oskodeshka.
Speaker 3 (10:28):
Thank you so much for your time.
Speaker 4 (10:29):
Seenior analyst at Swiss Code Now coming up bitcoin ooh,
nearing one hundred thousand dollars. The Trump dream is looking
towards creating a new crypto role. Well that next is
the Bluebeg technology Bitcoin's massive run. Well, let's see the
(10:59):
digital asset edging now towards one.
Speaker 3 (11:01):
Hundred thousand dollars.
Speaker 4 (11:03):
This as the Trunk team is said to be mulling
over creating its first ever crypto role that would oversee
cryptocurrency policy making. Marlbrenberg's Kaylee lines chows us Now, so
just what is.
Speaker 8 (11:13):
Being weighed right now? Well, a lot of it, Caroline
is still up in the area. This would be the
first crypto specific role in the White House, and in
many ways this would signal that Donald Trump was serious
about the crypto related promises he made on the campaign trail,
which of course included making the US the bitcoin capital
of the world, firing the current chair of the SEC,
Gary Gensler. He also had said he wanted to make
(11:35):
the first ever crypto presidential Advisory Council, So in some
ways it shouldn't be all too surprising that he is
looking at a role like this, But it's not abundantly
clear at this time what form this would actually take,
if this is going to be an actual White House
staff position or just someone who is more of a
crypto zar quote unquote if you will. I was speaking
to a source on this this morning who said, yes,
this is all very up in the air. They are
(11:57):
seriously looking at this, but it's just not clear what
kind of power would come win this role. If they
are going to have actual staff reporting into them, for example,
if they will just be advising, or if this really
ultimately is going to be about liaising and coordinating.
Speaker 3 (12:10):
And by coordinating I mean all the.
Speaker 8 (12:11):
Different parts of government that are involved in overseeing crypto here,
so between the White House and Congress, who would actually
write digital asset legislation, or the White House and other
agencies that have jurisdiction over this, like the CFTC and
the SEC. And I would just point you, Caroline to
the tweet from Kristin Smith, the CEO of the Blockchain Association,
yesterday she posted on x and she said, look, the
cryptos are idea quote unquote is a great idea, but
(12:33):
ultimately they can only actually do things if there is
a strong Secretary of the Treasury and SEC chair to
go along with that, and of course, Caroline, we have
not gotten picks for either of those roles from the
President elect yet, so a little.
Speaker 4 (12:46):
Of this is signaling, Kaylee. But just remind us on
the campaign trail, some of the promises to make the
US the crypto capital of the planet, to make mining
of a mole focus to the United States, to have
even well reserve of bitcoin. Is that necessary to have
congressmate moves or can this be done from an executive order.
Speaker 3 (13:07):
From a White House only led position.
Speaker 8 (13:10):
Some of it can in that the president obviously has
the power of appointment, so he could put in place
a more crypto friendly SEC and CFTC chair who would
be more lenient with the industry perhaps than we have
seen in the Biden administration. But a lot of this
will come down to Congress as well. It was actually
part of not just Donald Trump's campaign platform, but it
was in the Republican Party platform that was adopted at
(13:32):
the convention in Milwaukee this summer that they want to
end what they call democrats Unamerican crackdown on the industry.
So that would signal that this is kind of more
of a government wide effort than just what's going to
be coming out of the White House in Congress. Will
be really important here is they have the law making
power in terms of actually delineating jurisdiction between these different
agencies that kind of have been jockeying for control over it.
Speaker 3 (13:54):
And the chairmanship of.
Speaker 8 (13:55):
The House Financial Services Committee is something we should pay
attention to. It's going to be decided in the next
few months is one hundred and nineteenth Congress comes in.
Chairman currently Patrick mckenry is going to be leaving the
House and there's kind of a three way battle going
on between Congress and Bill Barr or excuse me, Andy
Barr of Kentucky, Bill Hezenga of Michigan, and French Hill
of Arkansas, who currently leads the subcommittee specifically on digital assets,
(14:18):
and so when where it's chairing, the Financial Services Committee
will have a lot of control over the shape of
the legislation that ultimately could work its way through Congress.
Speaker 3 (14:25):
And that's going to be.
Speaker 8 (14:26):
What a lot of the industry's lobbying time is spent
specifically on on.
Speaker 4 (14:31):
Industry lobbied going into these elections, and it works. I mean,
the money that was put to work. Backing crypto friendly
congress people is important.
Speaker 3 (14:40):
Right, absolutely.
Speaker 8 (14:42):
I mean you look at packs like fair Shape, for example,
that deployed tens of millions of dollars in some of
those key congressional races, including for example, a successful ousting
of the current chair of the Senate Banking Committee, Shared
Brown of Ohio. He was ousted not just by any
candidate but a crypto entrepreneur himself, and Bernie Marine the
incoming Republican Senate designate or Senate elect So that is
(15:05):
going to be really interesting to watch. As I mentioned
the House Financial Services Committee. The Senate, of course, is
going to be where a lot of the uphill climb
has been historically when it comes to advancing the interests
of the crypto industry. And Tim Scott will now be
taking over as the Senate Banking Chair.
Speaker 3 (15:19):
I would just point out.
Speaker 8 (15:20):
That the ranking member could very well be Democratic Senate
that Elizabeth Warren, who faced a crypto opponent of her own,
but one ultimately her reelection race, and she may have
a little bit of a role in at least effort
in attempt to pull back some of the efforts to
advance the industry's interest. But the crypto money that was
active in this election cycle.
Speaker 3 (15:38):
Has no sign that it's going anywhere.
Speaker 8 (15:40):
Now that they've proven that they actually can make a
difference in outcomes, here will We're watching to see how
this pack and other interest groups and the others individuals
who were contributing to it, think Mark and Dreson for example,
or Brian Armstrong, how active they will be in the
midterms in twenty twenty six and then ultimately in the
next presidential cycle in twenty twenty eight, always.
Speaker 3 (15:58):
Pushing us forward. Kaylee me, thank you.
Speaker 4 (16:01):
Now, let's talk about artificial super intelligence.
Speaker 3 (16:05):
Perhaps it's not far off.
Speaker 4 (16:06):
Last months off Bak CEO Masayoshi's son and predicted then
it would become a reality by twenty thirty five. Well,
former Google CEO Eric Schmidt, who is out with a
new book on artificial intelligence, says it might be coming
even sooner than we think.
Speaker 3 (16:19):
Take a listen.
Speaker 9 (16:20):
It's probable that we can build systems that are the
technical term is super intelligent, where you have a simple
system sarry a single system that is at the nineteh
percentile of physics, math, chemistry, and arts and so forth.
Speaker 7 (16:36):
No human can do that.
Speaker 9 (16:38):
It looks like these systems will be not only available
in the next five years because we already have examples
of passing these tests already, but also that they'll be
broadly available for all society.
Speaker 4 (16:49):
A bit more on AI app because happening today in
New York the Evident AI Symposium, where some of the
most senior leaders of the world's banking and artificial intelligence
areas came together. They're discussing the AI adoption in the
banking sector based on the latest data from an AI index.
It's a global standard benchmark for aimaturity in banking. I
sat down with JP Morgan Chase chief Data and Analytics
(17:11):
Officer Teresa heitsen Rada and asked if look, the efficiencies
from AI are the number one return on investment.
Speaker 3 (17:19):
Take listen.
Speaker 1 (17:21):
The magic question is what's what's the return?
Speaker 10 (17:23):
Right?
Speaker 1 (17:24):
And the answer is very, very difficult to quantify, And
I think at this stage of the game, the return
is a little bit of efficiency for a lot of
people every day, but not something that you can actually
hang a number on. So I think that the next
generation of where we see this heading is, rather than
just using it to help you write your email or
(17:44):
summarize a document, is when we start to plug the
technology into knowledge bases, when we start to actually look
at the workflow of a particular person or group and
linked together the steps in that workflow with the tools
that they need to be able to do their So
you basically go from the five minutes of efficiency to
the five hours of efficiency. That's a road like that's
(18:07):
going to take some time to actually get there, but
I think that's where we see it evolving. So there
are near term benefits, but I think the most compelling
is just putting the technology in people's hands and letting
them start to experiment and understand how it actually works
and how it can be used.
Speaker 4 (18:24):
JP Morgan Chase Chief Data and Analytics Officer Teresa heisen Weather.
There I mean, while coming up, it's official the DOJ
does want Google to divest its Chrome web browser.
Speaker 3 (18:34):
All details next.
Speaker 11 (18:34):
This is bring big technology.
Speaker 4 (18:50):
The Justice Department and a group of states do indeed
suggest that Google may have to divest its popular Chrome
web browser and imposed limits on its Android operating system
in order to address anti trust concerns, all confirming earlier
reports by Bloomberg.
Speaker 3 (19:04):
For Now, Bloombergsley and Nyland.
Speaker 4 (19:06):
Joins US and it was your reporting that ultimately has
been confirmed. Just tell us through how realistic still this
divestment would be.
Speaker 12 (19:16):
We'll see probably in early spring. The Justice Department has
said that they want Google to be forced to sell
off Chrome.
Speaker 3 (19:24):
It's not clear who would buy it.
Speaker 12 (19:25):
You know, the top two other browser makers, Microsoft and Apple,
who have the cash, the Justice Apartment probably wouldn't want
to let them, so the next you know, most obvious
candidates are probably people like open ai or Perplexity or
some of the other AI startups that would really love
to have this kind of an avenue to customers for
their AI search goals. Other things that the Justice Department
(19:51):
put in there is that they want Google to have
to license a lot of its data. That's probably a
more realistic thing that would have to happen, and and
it would definitely give a boost to other search engines
and possibly AI startups who would now.
Speaker 3 (20:05):
Not have to scrape the web themselves.
Speaker 4 (20:07):
Alphabet once again one of the key drags on the
NASTAC today. Investors worried about this alphabet itself, worried about
it coming out, and still once again trying to say
that this is a radical turn of events if imposed,
how complex is it to actually see a company sell
off such businesses? And how much do we have to
(20:27):
now wait for the legal maneuvering to play out? They
now have what a couple of months to respond.
Speaker 12 (20:33):
They have until next month to respond, and then the
Trump administration in March gets to have another reply. So
if the Trump administration wants to change this proposal a
little bit, they are allowed to. We probably won't have
a decision from the court until next summer.
Speaker 7 (20:48):
The judge has promised.
Speaker 12 (20:49):
That he will rule by August twenty twenty five, but
then Google says it will appeal. That'll take another year,
So at the earliest we're looking at them having to
do this in probably twenty twenty six. But I think
people are recognizing that Chrome is actually pretty important to
Google's business. It is, you know, an avenue through which
a lot of people access it's a search engine, but
(21:12):
it also is how Google collects a lot of the
data that underlies its advertising. You know, whenever a user
logs in, it keeps a lot of information about the
browser history.
Speaker 3 (21:23):
You know where.
Speaker 12 (21:24):
People are going online, what they're looking at, and then
Google uses that data to help sell advertising. So this
could have knock on effects to how some of Google's
core businesses.
Speaker 3 (21:34):
Me and Eilan appreciate the update. Thank you.
Speaker 4 (21:37):
Now it's time for talking tech, and first up, we're
talking PDD chares as you see plunging after warning that
it's profitability will trend downwards over time because of intensifying
competition in its home market. Of course, of China, PDD,
which competes with Ali Barba and its team, is struggling
to catch up with unspecified rivals because of the lack
of expertise, speaking of which Ali Baba has a pointed
(22:00):
better and executive Jeng Fan to oversee its entire online
e commerce operation. Now Jiang will lead a newly created
department that consolidates Ali Baba's online shopping assets, making him
the most powerful person in the company after the CEO
and chairman and Baido. It recorded its biggest revenue drop
more than two years after China's economic malaise. Under mind
it's push into general to AI. Now the company's ernie
(22:21):
Bot has fallen behind Bite Dances d'albao in China usage,
while its core business is losing ground. Two newer social
platforms off by seven percents.
Speaker 3 (22:30):
You'll see now coming up.
Speaker 4 (22:31):
We're going back to the US earning story Nvidia the
impact on the chip sector at large. We're currently off
by one point three percent, one point two percent for
the world's most valuable publicly traded company. This is the
room Mate Technology. Welcome back to Blue meag Technology and
(22:57):
Karin Hider, New York. Let's get back to it. In
Video's earnings. Here is the reaction from just some of
our guest somebody Meg Television.
Speaker 3 (23:05):
I have a feeling that we've reached peak Navidio. How
long can you stay perfect? Great?
Speaker 13 (23:09):
So I think that, yeah, I mean, I do think
that a lot of investors are expecting a little bit
too much.
Speaker 10 (23:16):
There's a little bit of this kind of fatigue.
Speaker 7 (23:18):
But make a mistake. This was a remarkably good quarter.
Speaker 3 (23:22):
When they're putting up in terms of Hopper, in terms
of overall growth, it's exact what the what the bulls want.
Speaker 13 (23:29):
Never underestimate Wall Street's ability to miss the bigger picture.
This company has double the operating margin of any other
Meg seven outside of Microsoft. To nitpick in Video on
those margins is truly missing the big picture.
Speaker 3 (23:43):
Let's nitpick some more.
Speaker 4 (23:44):
Daniel Pelling, portfolio manager and senior technology research analyst at
Sans Capital joins us now and Daniel, Yeah, we're quibbling
on a margin of seventy five percent going to somewhere
to seventy three to seventy four percent, so much more
profitable than rivals.
Speaker 14 (24:00):
Yes, Hi, good morning. Yeah, so well, I think actually
the gross Martin stuff is largely behind us. They've guided
for sort of seventy five percent exiting the year, And
I know this was a concern a while back, but
I would actually argue it's somewhat frankly irrelevant. I think
what was much more important was two things. First of all,
(24:20):
the company noted that the scaling laws are well and
alive and that we can both scale training and inference
going forward. So this entire discussion about how we're hitting
a wall is sort of probably off, and there's just
many walls and they're far away. And then the second
really important thing I thought that they mentioned was is
to say that data centers are becoming a little bit
like smartphones, i e. They are electricity constrained. And what
(24:44):
that means is just amazing, right, because that means every
year the hyperscalers have to buy a new chip because
the new chip is better in terms of electricity usage,
and we've never had that before in the data center.
So those two things I think are very positive for
the medium and long term of the company.
Speaker 4 (24:59):
I want to go about to the scaling laws, Daniel,
because yes, we've had Jensen Wang actually back on the
conference trail just earlier this week talking about how compute
power is going to continue to extend, and indeed the
more data, the more compute you have, the more sophisticated.
Speaker 3 (25:12):
The larger language models will become.
Speaker 4 (25:14):
But then you listen to the reporting out of Bloomberg
around open ai and the latest iteration of large language models.
Just stuttering a little, Why are you so confident that
scaling laws.
Speaker 3 (25:23):
Aresketden in place?
Speaker 14 (25:25):
Yes, So, I think the key thing to note here
is that all that information that we have is based
off of smaller clusters. So now we for the first
time have a cluster that's something like one hundred or
two hundred thousand GPUs that is owned by Elon Musk
and we'll see what comes out of that in the
next twelve months. And secondly, and more importantly, maybe the
(25:47):
black clusters start at one hundred thousand GPUs. Hopper started
at ten thousand GPUs, so we'll have to wait the
next twelve months. What can we train with these much
better chips and much larger clusters and whatever out of that?
That is the crux. So I think most of the
stuff that's in these articles is kind of backward looking
because we just didn't have the chips nor these big
clusters yet.
Speaker 4 (26:08):
That has been reporting about the latest valuation of Xai
and the fact that they're raising money simply to buy
more GPUs, But there is that worry about density of
customer for n video. Are we going to see a
broadening out?
Speaker 14 (26:23):
Yeah, So I think the big question here is like this,
and I think you're right. So the big customers are hyperscalers,
venture finers, startups, and actually healthcare companies are using training
a lot too. My sense is that the most important
thing to look at here is for the next year
is agentic AI and this idea of that we're spending
much more compute on inference and letting inference think. And
(26:44):
I think it's starting just now, and if we can
see this, then companies can really start using this via
using service now, Workday, Microsoft, et cetera, et cetera. So
I think that might be a big tipping point and
we're literally at inning one of that. And just to
be clear, agentic AI is this sort of THISI they
can do things autonomously within an enterprise and help with workflows.
Speaker 4 (27:05):
We've had a lot about agents from the likes you
mentioned service now we've also heard it from Salesforce.
Speaker 3 (27:09):
And the light Daniel.
Speaker 4 (27:10):
So when we look at your funds managed what the
top holding is in Nvidia, does that start to broaden
out for you or do you mean comfortable with your
alignment on in video even though it's at what the
most valuable market capitalization in the entire world.
Speaker 7 (27:27):
Two things.
Speaker 14 (27:28):
One, so I think there's a good scenario to be
made that Nvidia becomes the first ten trillion dollar market company.
And I'll tell you why. I mean, even if you
just look at the top four hyperscalers, they're estimated to
generate about three hundred billion dollars of free cash flow
next year alone. And what I'm trying to say is
there's a lot of dollars left to go to Nvidia
(27:49):
if the scaling.
Speaker 7 (27:50):
Laws hold true.
Speaker 14 (27:51):
So that's one, we're very bullish on video or continue
to be positive on video. But two, the trends are
really proliferating into other spaces, and maybe I'll make two
comments there. First, First of all, the software company Service
now Microsoft et Centa, they will benefit. But there's also
other companies you know that have significant competitive advantages and
it can really take AI and monetize it right. And
one great example might be a company called Axon that
(28:14):
you know, manufacturers tasers and cameras for the police force.
They've added an ability to transcribe you know, police when
they're engaged with people, and that says forty percent of
time for that police officer. That's a huge improvement, and
the stock has started to price them in as an
AI winner, and I think you'll see more of that
over time too.
Speaker 4 (28:32):
Another company for us to focus in on for our
audience standing pelling. So great to have time with the
senior technology research analyst, Sam's Capital. How interesting that he
kept referencing inference. We're going to have a conversation on
that now because we have a perspective and someone in
the field. Rodrigo leac CEO of sa Manova is an
AI infrastructure company that delivers advanced AI solutions for enterprises.
You're kind of an Nvidia competitor upstart.
Speaker 3 (28:56):
And I just want to articulate for our audience what
it is that you do. Yeah, no, thanks for having us.
Speaker 15 (29:00):
So some of the is started as a startup coming
out of Stanford University and palto really focus on efficient computing.
And so as we're seeing in video scale and you
see all the all the great numbers that that people
are talking about with video or we're now seeing, is
what happens when the world wants to go into production
and going to scale. And one of the things that Franklin,
we're not talking enough about is the fact that power
(29:22):
and energy is going to be a big bottleneck, and
you know, people are talking about building nuclear power plants
and things like that. Somanova Te took the focus of
how do we actually take AI at scale power efficiency
and drive it down so that you can actually scale
without running into these limits that we start seeing with
both companies and countries running out of power and figuring
(29:44):
out how to actually provide enough energy to supply these GPUs.
Speaker 4 (29:49):
So your view is you can't like there's going to
be a certain limitation on the amount of energy efficiency
that a company like and Video is ever going to hit.
With the GPU whereas you of a lot.
Speaker 15 (30:01):
Well, so today, for example, you know, a single rack
of Nvidia is north of one hundred kilowatts, and just
to give your audience a sense, that's one hundred, it's
one hundred homes that it powers. And so if you
think about someone of technology, we're able to drive ten
next to performance at one tenth the power now because
of a completely different architecture. As you know with n video,
(30:22):
it's an architecture that's been around for several decades and
so we're trying to evolve this graphics GPU architecture into
AI enabled now for inferencing, there's so many new techniques
that you can actually apply because you don't need to
do all the heavy computing that training requires, and so
you're able to drive a much much lower footprint in
terms of power, in terms of costs, in terms of space,
(30:44):
which ultimately is required when you're actually going to scale.
Speaker 7 (30:48):
And my belief is AI.
Speaker 15 (30:50):
Inferencing is going to be ten times more of a
footprint than training, and so I think we're going to
have to think about how to be energy efficient, how
to be cost efficient before it gets to the scale
that everybody get.
Speaker 2 (31:00):
To have it.
Speaker 4 (31:01):
Can you for our audience that is not only an
investing audience, but really a technologically savvy audience. Describe to
us what it is that's different that enables this higher
energy efficiency within your particular.
Speaker 3 (31:13):
Type of technology, because it's not GPU at all.
Speaker 15 (31:15):
What is it, right, It's not graphics and so it's
something that was inventing researchers stand for University. It's called
data flow architecture, and what it does it allows you
to actually map the hardware in the way that a
neural net wants to operate. And so with n video today,
you're having to break up these neural nets into these
cores that exists inside the graphics chip, which turns out
to be a lot of work, a lot of energy,
(31:36):
a lot of power salmon over with the data flow
architecture allows you to avoid all of that extra power
that's be incurred when you actually break up these neural
nets and let it flow through the machine, and so
you're getting these order magnitude improvements in power efficiency, which
ultimately allows you to be able to deploy these in
existing data centers, which, by the way, is one of
the things that I think is not talked about. When
(31:58):
we actually acquire GPUs because you also have to upgrade
the data centers where for our technology going into existing
data centers with existing power grids, with existing cooling systems,
it becomes one of those things that we allows us
to generate value for enterprises really really quickly.
Speaker 3 (32:15):
What enterprises you already locked in with?
Speaker 4 (32:17):
How are you managing to sell and this narrative into
corporates and federal government?
Speaker 15 (32:22):
Yeah, we're yeah, we're as a startup again and videos
at the top of the tier. We're a startup that's
you know, coming in and then providing disruptive technology. We're
the most deployed in the US government today as a startup.
Speaker 3 (32:34):
You know, we're going to go into national abs.
Speaker 7 (32:36):
And places like that.
Speaker 15 (32:37):
And and today we power also Saudi Aramco's Metabrane, you know,
so they're internal GPT. This is a private GPT with
data trained into it and now operating air gap within
the company. And so companies that worried about their data
their secured their privacy can now train these models on
assemblent of platform completely securely and then deploy within their
(32:57):
own infrastructure because you don't have to expose the.
Speaker 3 (32:59):
Data outside your firewalls.
Speaker 4 (33:02):
There are other startups in this space, or indeed trying
to offer more energy efficient options and pay for example
also back by Oracle. You yourself used to work to Oracle.
I'm interested as to what your next step is to
really gain market share.
Speaker 3 (33:17):
Is it about raising funds?
Speaker 4 (33:18):
I know you've got big VC backing for example.
Speaker 7 (33:21):
Yeah, I think you know.
Speaker 3 (33:22):
It's really good point.
Speaker 15 (33:23):
My belief when we started the company back seven years.
Speaker 3 (33:26):
Ago is about access.
Speaker 15 (33:27):
When the marketized AI, you got to make it available
to everybody. As your previous guests just said, is today
in video is really in large scale only to a
few customers. What we want to do by collapsing the
power is to be able to make it available to everybody,
to make it easily available, and that means lower cost,
much lower power and then able to deploy in very
(33:51):
tight spaces and not regarding needing and giggle what data center.
So some of that we focus very much on completely
restructuring the power, the power requirements of the AI infrastructure
so they can deploy racks at performance that much higher
than in VideA, not less than VideA, but much higher
than nvidea and much lower power.
Speaker 4 (34:11):
Samonova Rodrigo Lean CEO talking about some of the alternatives
out there. John and video, We thank you for coming on. Meanwhile,
coming up Palo Alto Networks earning. They are looking at
the first quarter that is proving itself with this focus
on platformization. We'll dig into that word in a moment.
CEO Niqushaurra joins us. This is Blue Bag Technology, cybersecurity company,
(34:45):
Palo Alto Networks, releasing earnings after the market Wednesday. And look,
we're seeing revenue climbing fourteen percent, earnings climbing thirteen percent.
We're seeing a focus on platformization and also a stock split.
We're currently up one and a quarter percent. But let's
get to the details with the CEO Nikosh Aurora of
Palo Alto Networks without some volatile trading for the market
now pushing you higher. And I'm just interested on the
(35:07):
nu once you are trying to tell.
Speaker 3 (35:09):
A different story to your investors.
Speaker 4 (35:10):
You're talking about platformization, you're talking about bundling. But still
they want to turn to your billings number. How do
you manage to get that across to them that stopped
looking at it because there was a little bit of
worry about billings.
Speaker 2 (35:21):
Well, first of all, thank you for having me, Carol,
and I appreciate you saying the word platformization, because this
is something we've been trying.
Speaker 7 (35:27):
To tell the market.
Speaker 2 (35:28):
I think we are seeing a c shift in the
cybersecurity industry where people are tired of buying point products
which are not solving the problem. Today, you have to
make these products work together. And I was listening to
the conversations before this. Everybody's talking about AI. AI in
our business means things are going to happen faster. Attacks
are going to happen faster. You're going to have to
(35:48):
resolve them faster, which means you need a common language,
you need common data which actually points towards integration and
stuff that works together, or points towards platformization.
Speaker 4 (36:00):
Cloud platform strategy that seems to be earning you plenty
of companies.
Speaker 3 (36:04):
What is it accumulative?
Speaker 4 (36:05):
You've got one, one hundred platformizations?
Speaker 3 (36:09):
Ultimately, is that more valuable? Is that more reven generating?
Speaker 4 (36:12):
Because people are worried that maybe within this bundling there
is some discounting going on.
Speaker 2 (36:17):
Well, I think it's important to understand once you deploy
a platform and it works well for you, you're not going
to replace it. So You're coming from an industry where
you look, we've had companies that have worked well for
eight or ten years, but they've been through a cycle,
and then you have to replace them and replace them
on a net new set of vendors because the last
ones didn't stay innovative enough. And what we're trying to
do is we're trying to stay evergreen. We're trying to
(36:38):
be innovative. We're trying to make sure we embed the
platform as part of your security solution that you logically
come to us for the next piece of innovation. And
you have to go through a huge process of constantly
transforming your security. In state a state today, more customers
have security transformation projects on tap than any other technology
transformation because the security landscape is constantly changing.
Speaker 4 (37:00):
Talk about the cyber landscape and whether it changes with
a new administration. Is there going to be more consolidation,
and is there going to be more deal making from
you look, I get.
Speaker 2 (37:10):
Past that in different ways. You know, I'm watching it
like everybody else and saying it sounds like we're going
to see the new doge go out and do a
lot of cost cutting and deploying a lot of technology.
Now Traditionally, the federal government in the United States has
been sloughed off new technology rightfully so, perhaps because they
want to sense their way through it. But as you
see efficiency, as you see technology investment happening, it has
(37:33):
to be secure. So I think security gets a knock
on benefit if we actually see a technology wave coming
from an administrative perspective, Do I expect regulations to get easier?
Possibly from an M and A perspective, Do I expect
that things.
Speaker 7 (37:46):
Will move faster?
Speaker 3 (37:46):
Hopefully? So, I think generally it should.
Speaker 2 (37:49):
Sort of board well for all of us in technology,
But you know, we'll all find out.
Speaker 1 (37:55):
We'll wait and watch.
Speaker 4 (37:56):
Look as you innovate, as particularly within the world of
gender to AI, we managed to get more sophisticated cyber protection,
you're also getting more sophisticated hackers. The US government has
just flags and flaws with Palo Alto's owned products that might.
Speaker 3 (38:11):
Have been exploited by hackers.
Speaker 4 (38:12):
Can you just tell us a little bit about whether
that was a realistic one for.
Speaker 3 (38:16):
You, what patches have been made? Well, like what.
Speaker 2 (38:19):
Happens is because we're coming from a hardware industry where
we sell products, customers deploy them, there's a long product cycle,
and you've seen this across the industry. At any point
in time, you know, there's always some bug that is
out there that needs to be passed and fixed. So
we're very good about making sure we work with our customers.
We have full visibility into who has what product. We're
able to work with them to pass those things. I
(38:40):
think as we move from a hardware based solution business
to a software business, which the world is moving to,
that becomes a lot easier to fix. I think software
beats hardware from a security recency perspective all the time,
So as the world transforms there we'll see less than
less of that in the future.
Speaker 4 (38:57):
As the world transforms, everyone becomes an investor retail community
as well. Talk to us about your two for one
stock split. You're already up what thirty percent on the year.
Is that why you spread it out? Is it about
giving more access?
Speaker 2 (39:09):
Well, we want to make sure that there is ample
access to our stock to retailed investors. You know, five
years ago when I started, we were a sort of
a niche name from the cybersecurity industry. Or delighted that
we are now part of the general conversation. We want
to make sure everybody has access to our stock.
Speaker 4 (39:22):
Well for now, we'll currently see it's going to be
action in a two for one stock split. We'll see
how much adoption there is around that. But in cash
Aurora on the fundamentals and some of the share moves.
CEO of pallel to Networks, we appreciate your time today.
Speaker 3 (39:35):
Thank you.
Speaker 4 (39:36):
Coming up more earnings, Snowflake giving a pretty strong outlook
for its product sales growth.
Speaker 3 (39:40):
We dig in this is.
Speaker 11 (39:41):
Blue meg Technology.
Speaker 3 (39:55):
More earnings for you.
Speaker 4 (39:56):
Snowflake shares absolutely jumping after the company go some pretty
solid sales outlooks for us, suggesting newly launched products are
receiving a strong reception from customers. Brady Ford has the
inside take, I mean extraordinary move to the upside from Snowflake.
We know it's a volatile stop, but what's managing to
win over customers?
Speaker 10 (40:14):
Bernie in software right now, the big question is who
gets left.
Speaker 7 (40:19):
Behind in the age of Genai.
Speaker 10 (40:21):
For a long time, investors worried that Snowflake was focusing
on maybe a two narrow slice of the data market.
You know, the services used to kind of organize and
analyze all your data. And Snowflake is successfully making an
argument that no, in the era of Genai, when folks
need to understand such a wide array of data and
(40:42):
turn it into interesting applications, we will remain a platform
where they spend their money and.
Speaker 7 (40:48):
Make their decisions.
Speaker 10 (40:49):
And so that's what investors are reacting to in these results.
Speaker 4 (40:52):
They organize, they analyze your data, but when it comes
to unstructured data, they're having.
Speaker 3 (40:56):
To make purchases. Here, it's a bit of.
Speaker 7 (40:58):
M and a Yes.
Speaker 10 (41:00):
So many have heard of this famous rivalry between Snowflake
and Data Bricks for a long time. The argument was
that data Bricks handled the unstructured data better, and that's
the real you know, something that's not in a spreadsheet,
something that's real disorganized. Maybe it's a bunch of Google docs.
You want to be able to scan through and say, hey,
give me insights. And so Snowflake has worked to beef
(41:21):
up their ability to deal with that kind of hard
to access data. And yeah, last night we saw them
make an acquisition of a startup that focuses specifically on
ingesting this kind of messy, unstructured data you need for
a lot of Genai use cases.
Speaker 4 (41:37):
I know that you've reported a lot on that rivalry.
So see if that M and A does indeed help.
But under the guidance of the new leadership in Ramaswami,
he's also trying to integrate large language models, right, how
is this being seen in the anthropic deal Exactly?
Speaker 10 (41:52):
They want to bring more LLLM access right into the
Snowflake platform. So if you want to do things like
against stand through all your information and pull out insights,
make dashboards, make visualizations, you can do it right in Snowflake.
You don't have to go open up a new you know,
Azure instance or leave the platform. So bringing anthropic models
(42:14):
right into Snowflake is again hoping to make it a
platform where you remain where kind of your most important
data is. They really hope to be a center of
gravity for all your information and not just something you
you know, use when you want to make an analysis
of a spreadsheet somebody sent to you.
Speaker 4 (42:33):
Brady Ford, across the Snowflake moves and indeed some of
the intricacies and the earnings we thank you. Look, let's
just get a border perspective of how snowflakes thirty percent
the upside fits into what's happening more broadly across the
NASDAK because look, it hasn't been an round out risk
on kind of a day. In fact, we've been lower pulled,
lowered by about two tens a percent, large part because
(42:54):
of Nvideos off by one point seven percent, Alphabet two
on the back of that formalization of Bloomberg reporting. Then
indeed the DOJ does want it to be selling off
Chrome for example, but in video of by one point
seven percent is important because it is the most valuable company.
It is going to have a huge points perspective when
it drags lower.
Speaker 3 (43:10):
But look, we were.
Speaker 4 (43:11):
Anticipating in the options market there's mure an eight percent
swing upside or downside. So this is relatively stable, shall
we say. After earnings did show a ninety four percent
increase in revenue, a seventy percent increase in revenue for
the Guide, and margins being pretty strong.
Speaker 3 (43:26):
Interestingly, the chops that the.
Speaker 4 (43:28):
Socks the chip sector is up some seven percent, so
the rest of the market doing well today. Have a
look at what's happening with Alphabet because the parent company
of Google off another six point eight percent today. Once again,
this anxiety surrounding antitrust and how it might have to
be split off weighing on the stock that does it.
In this edition of Bloomberg Technology, a lot to get through.
(43:49):
You want to check it all out again on the podcast.
You'll find it on the terminal, as well as online.
Speaker 3 (43:53):
On Apple, Spotify, and iHeart. This is Bloomberg
Speaker 7 (44:00):
Today, The Healthy Everyday, Healthy Everyday, The health of
From the heart where Innovation, money and power Collie in
Silicon Valley, NBN.
Speaker 2 (00:07):
This is Bloomberg Technology with Caroline Hyde and Ed lud.
Speaker 3 (00:10):
Love live from New York.
Speaker 4 (00:25):
This is Bloomberg Technology coming up in video.
Speaker 3 (00:28):
Says it's new.
Speaker 4 (00:28):
Chipper on track, but the rush to get the product
lineup out the door is.
Speaker 3 (00:32):
Proving more costly than expected.
Speaker 4 (00:34):
Plus, Bitcoin rises to new heights as the Trump transition
team weighs the creation of a dedicated white house cryptopost
and the CEO of Palo Alto Networks joins on the
heels of its earnings.
Speaker 3 (00:45):
We'll discuss how AI can.
Speaker 4 (00:47):
Help stop the threats of the future, but first we
get straight to the world of in video and indeed
the points trag that it's having on the Nasdaq is
somewhat We're off by eighteen percent, but it's been a
volatile day of trade as people.
Speaker 3 (00:59):
Try to pass.
Speaker 4 (01:00):
Ultimately, how strong future growth still looks under in video
and indeed the supply side headaches that much has been discussed.
Let's get straight to within King and look, we'd all
been fretting about Blackwell the product lineup, how easily it
is to get into customers. Hans did Jensen Wong manage
to calm investors' nerves?
Speaker 5 (01:19):
I mean, obviously to an extent. We haven't seen the
massive fall off that we've seen manifest previously with concerns.
He talked about, Yeah, we're getting it to market. We're
actually going to be probably getting more to market than
we had thought we would. At the same time, it's
going to cost us.
Speaker 4 (01:36):
Yeah, let's just weigh in on the profit margins, because
you do such a great job at giving the context here. Ultimately,
this is a company that gave us ninety four percent
increase in revenue in the third courting guided to another
seventy percent increase, give or take plus or minus two percent.
But the profitability is what just coming down ever so
slightly to what seventy three to seventy four percent margin?
Speaker 5 (01:57):
That's right, I mean, I think that's the right way
to look at it. They've set such high expectations, right,
they were at seventy five percent, which had narrowed slightly.
But seventy five percent is like software mhudgeons, Right, this
is a semiconductor company, right, you know, AMD's twenty points
south of that. You know, Intel isn't even in the
same zip code right now. So yes, they're losing a
(02:18):
few points because they're spending more in engineering to get
those chips out, get those systems in place quicker.
Speaker 2 (02:24):
But really this is a very very high bar.
Speaker 4 (02:27):
And then full production comes what the back end of
their fiscal twenty twenty five in. But I suppose they
still have to address the issue that. Sure they're getting
them out the door, but it's to the same kind
of companies. How are they managing to diversify their demand base?
Speaker 5 (02:43):
Well, I mean that was one of the things that
we saw in the release. They're not right. The amount
of revenue coming from those big companies, the Microsoft's, the
Amazons has actually gone up from forty five percent to
fifty percent, So like half of the most important divisi
that they have is dependent upon just a few companies.
Investors are definitely concerned about that. Why they're concerned about
(03:05):
that because these companies are making their own chips, trying
in effect to replace in video and everybody else that
supplies them.
Speaker 4 (03:13):
Is there any real competition here, because we've heard it
time and time again that as much as AMD tries,
as much as Intel tries, really in videos, the any
game in time for here.
Speaker 5 (03:23):
Yeah, I mean that that definitely appears to be the case.
I mean that any questions about and demand, any questions
about losing market share were just irrelevant, right. I mean
they talked about incredible demand, demand that they cannot meet
right now. So you know that part of the quote,
the story is really in the future, if at all, Ian.
Speaker 4 (03:45):
King very much in the present when it comes to
the market move, We really appreciate it.
Speaker 3 (03:50):
Thank you.
Speaker 4 (03:51):
Let's get you more contact and what it means to
the rest of the market. Epshkere is with US Senior
and a list at Swiss Quote and Ibet. We turned
to you for the market implications because we were anticipating
as much as a three hundred billion dollar market cap
swing up to eight percent move higher or lower. Actually,
a lack of holatility must be a slight sign of relief.
Speaker 6 (04:14):
Well, actually it is because there has been a few
red flags in yesterday's report. As A was just talking
about the pressure on the productivity levels due to the
manufacturing challenges for the black Bell chip is one of them.
The fact that the big technology companies make up to
fifty percent of Nvidia's revenue is another sticky point. Yet
(04:34):
the results have been strong and despite these small red
flags that we got yesterday from the report, well, the aftermass,
the immediate selloff has been only twenty half percent, and
this morning we don't see that the Nvidia shares are
being sold at the level that we would expect them
to with such a disappointment, or if with the small
(04:55):
disappointment from investors, we would think that any miss that.
Speaker 7 (04:59):
Would be a excel off.
Speaker 6 (05:00):
Yet we just see that the market is quite resilient.
Speaker 4 (05:05):
We've got a bit too used to him managing to
smash expectations.
Speaker 3 (05:09):
That he had set and that of the market.
Speaker 4 (05:11):
And look, you only beat them slightly when it comes
to their own internal guidance IPEC. The rest of the market, though,
had been anticipating, well, maybe a questioning around valuation. What
do you make of the fact that we're still up
let's call it two hundred percent over the course of
this year. It's an extraordinary run up for one particular company.
Speaker 3 (05:31):
Can it be sustained?
Speaker 6 (05:33):
Well, it is absolutely an extraordinary run up, and now
we are questioning the valuation levels, especially if.
Speaker 7 (05:40):
The demand started to solve.
Speaker 6 (05:42):
This is not the case right now, but at some
point we expect that the demand from the big technology
companies will start swing, and because this is a big
chunk of the company's revenue, that's going to be a headache.
Speaker 7 (05:53):
There's another thing.
Speaker 6 (05:55):
Right now, the competition is not a problem because Nvidia
has got the most expensive but the most premium chips.
They've got the best performance, and the big technology companies
are looking for those. But if the demand was to
slow for the big technology companies, then Nvidia will have
to find other customers. And these other customers may not
be looking for the premium solutions. They will be just
(06:16):
looking for the solutions to manage their day to day
businesses and increase their productivity and increase their costs. But
they would also up for maybe more cost efficient and
well slightly more affordable chip solutions than the Nvida offers.
And I think that that's going to be a major issue.
Speaker 4 (06:35):
We're going to be talking exactly that later in the programming.
For now, when we think about the context of in video,
about twenty percent of all the run up in the
S and P five hundred is thanks to this one
particular name this year, about twenty five percent of all
earnings per share increase. How key man risk is in
video and how does one protect themselves against that.
Speaker 6 (06:56):
Well, obviously, the AI rally and especially has been one
of the major pillars of the market rally in the
US that we have seen over the past two years.
So anything any lack of epissites for Nvidia from now
on is going to have a negative impact on the
market sentiment. Now, we have seen that AI rarely broaden
(07:18):
toward the other sectors, other non technology sectors. But it's
worth noting that this company has added so much upside
pressure to the SMP five hundred, So anything less than
any kind of loss of momentum here will obviously be
well quite negative for the SMP five hundred, especially knowing
that the SMP five hundred today is struggling near its
(07:41):
all time high levels.
Speaker 4 (07:43):
So when you're sat ultimately in Europe and you're thinking
about a global perspective, it should people be broadening out
when it comes to just magnificent seven, let alone just
in video litlone just US bets on technology.
Speaker 6 (07:57):
Well, when it comes to technology, I really believe that
the US is still in a very dominant position, So
you wouldn't be going to to Europe for technology solutions.
There's one place that we actually like, and that's Japan,
especially given that the government there is going to give
some more support to their technology industry.
Speaker 7 (08:15):
So in terms of geographical.
Speaker 6 (08:17):
Diversification, Japan could be a good solution. There is China
as well that we have been looking at, but we
are still not positive for China because well, the chipwoar
has weighed greatly on sentiment and on the progress of
the technology advancement there, So we're not really looking at
China right now. So one place that investors could be
(08:38):
looking at would be Japan, but other than that, the
US is still in a very dominant position. So if
you want to invest in tech, well you would just
the first place you would like to go is the
US because this is where you have massive upside potential.
Speaker 4 (08:50):
So back the clients that are calling you, are they
saying they want to add to Invidia, for example, as
a name at these particular levels or do they just
keep their exposure where it is well.
Speaker 6 (09:01):
To be perfectly honest with you, I don't have clients
who would like to sell Nvidia, but at the current
levels people are more skeptical about entering into.
Speaker 7 (09:09):
The NVDA shares.
Speaker 6 (09:10):
What I mostly here is what level would be a
good entry level if we start seeing a price pullback
from the actual level, So people will still be buying
MVDA shares from this point on, but they're still at
disvaluations at this point in time, they will be looking
for good entry levels.
Speaker 4 (09:29):
Is there other names, smaller chip rivals that you've thought
would be a better entry point.
Speaker 6 (09:37):
Well, I think that AMD is at a good position
right now. Obviously, they do not offer the same potential
and d do not offer the same margins than Nvidia does,
but it's share price have come down quite significantly to
offer an interesting entry point at the current levels. It's
also to be said that AMD is about to launch
(09:58):
two powerful chips, one for the end of this year
and one for next year. And what's interesting with AMD
is they may not be as powerful as premium as
Nvidia chips, but they're clearly going to be more cost
efficient and there are many many sectors and companies out
there that will be looking for more cost efficient and
more affordable solutions, and AMD could actually help fulfilling that gap.
Speaker 4 (10:22):
MD currently trading at forty one times future earnings compared
to about a fifty times for a video effect Oskodeshka.
Speaker 3 (10:28):
Thank you so much for your time.
Speaker 4 (10:29):
Seenior analyst at Swiss Code Now coming up bitcoin ooh,
nearing one hundred thousand dollars. The Trump dream is looking
towards creating a new crypto role. Well that next is
the Bluebeg technology Bitcoin's massive run. Well, let's see the
(10:59):
digital asset edging now towards one.
Speaker 3 (11:01):
Hundred thousand dollars.
Speaker 4 (11:03):
This as the Trunk team is said to be mulling
over creating its first ever crypto role that would oversee
cryptocurrency policy making. Marlbrenberg's Kaylee lines chows us Now, so
just what is.
Speaker 8 (11:13):
Being weighed right now? Well, a lot of it, Caroline
is still up in the area. This would be the
first crypto specific role in the White House, and in
many ways this would signal that Donald Trump was serious
about the crypto related promises he made on the campaign trail,
which of course included making the US the bitcoin capital
of the world, firing the current chair of the SEC,
Gary Gensler. He also had said he wanted to make
(11:35):
the first ever crypto presidential Advisory Council, So in some
ways it shouldn't be all too surprising that he is
looking at a role like this, But it's not abundantly
clear at this time what form this would actually take,
if this is going to be an actual White House
staff position or just someone who is more of a
crypto zar quote unquote if you will. I was speaking
to a source on this this morning who said, yes,
this is all very up in the air. They are
(11:57):
seriously looking at this, but it's just not clear what
kind of power would come win this role. If they
are going to have actual staff reporting into them, for example,
if they will just be advising, or if this really
ultimately is going to be about liaising and coordinating.
Speaker 3 (12:10):
And by coordinating I mean all the.
Speaker 8 (12:11):
Different parts of government that are involved in overseeing crypto here,
so between the White House and Congress, who would actually
write digital asset legislation, or the White House and other
agencies that have jurisdiction over this, like the CFTC and
the SEC. And I would just point you, Caroline to
the tweet from Kristin Smith, the CEO of the Blockchain Association,
yesterday she posted on x and she said, look, the
cryptos are idea quote unquote is a great idea, but
(12:33):
ultimately they can only actually do things if there is
a strong Secretary of the Treasury and SEC chair to
go along with that, and of course, Caroline, we have
not gotten picks for either of those roles from the
President elect yet, so a little.
Speaker 4 (12:46):
Of this is signaling, Kaylee. But just remind us on
the campaign trail, some of the promises to make the
US the crypto capital of the planet, to make mining
of a mole focus to the United States, to have
even well reserve of bitcoin. Is that necessary to have
congressmate moves or can this be done from an executive order.
Speaker 3 (13:07):
From a White House only led position.
Speaker 8 (13:10):
Some of it can in that the president obviously has
the power of appointment, so he could put in place
a more crypto friendly SEC and CFTC chair who would
be more lenient with the industry perhaps than we have
seen in the Biden administration. But a lot of this
will come down to Congress as well. It was actually
part of not just Donald Trump's campaign platform, but it
was in the Republican Party platform that was adopted at
(13:32):
the convention in Milwaukee this summer that they want to
end what they call democrats Unamerican crackdown on the industry.
So that would signal that this is kind of more
of a government wide effort than just what's going to
be coming out of the White House in Congress. Will
be really important here is they have the law making
power in terms of actually delineating jurisdiction between these different
agencies that kind of have been jockeying for control over it.
Speaker 3 (13:54):
And the chairmanship of.
Speaker 8 (13:55):
The House Financial Services Committee is something we should pay
attention to. It's going to be decided in the next
few months is one hundred and nineteenth Congress comes in.
Chairman currently Patrick mckenry is going to be leaving the
House and there's kind of a three way battle going
on between Congress and Bill Barr or excuse me, Andy
Barr of Kentucky, Bill Hezenga of Michigan, and French Hill
of Arkansas, who currently leads the subcommittee specifically on digital assets,
(14:18):
and so when where it's chairing, the Financial Services Committee
will have a lot of control over the shape of
the legislation that ultimately could work its way through Congress.
Speaker 3 (14:25):
And that's going to be.
Speaker 8 (14:26):
What a lot of the industry's lobbying time is spent
specifically on on.
Speaker 4 (14:31):
Industry lobbied going into these elections, and it works. I mean,
the money that was put to work. Backing crypto friendly
congress people is important.
Speaker 3 (14:40):
Right, absolutely.
Speaker 8 (14:42):
I mean you look at packs like fair Shape, for example,
that deployed tens of millions of dollars in some of
those key congressional races, including for example, a successful ousting
of the current chair of the Senate Banking Committee, Shared
Brown of Ohio. He was ousted not just by any
candidate but a crypto entrepreneur himself, and Bernie Marine the
incoming Republican Senate designate or Senate elect So that is
(15:05):
going to be really interesting to watch. As I mentioned
the House Financial Services Committee. The Senate, of course, is
going to be where a lot of the uphill climb
has been historically when it comes to advancing the interests
of the crypto industry. And Tim Scott will now be
taking over as the Senate Banking Chair.
Speaker 3 (15:19):
I would just point out.
Speaker 8 (15:20):
That the ranking member could very well be Democratic Senate
that Elizabeth Warren, who faced a crypto opponent of her own,
but one ultimately her reelection race, and she may have
a little bit of a role in at least effort
in attempt to pull back some of the efforts to
advance the industry's interest. But the crypto money that was
active in this election cycle.
Speaker 3 (15:38):
Has no sign that it's going anywhere.
Speaker 8 (15:40):
Now that they've proven that they actually can make a
difference in outcomes, here will We're watching to see how
this pack and other interest groups and the others individuals
who were contributing to it, think Mark and Dreson for example,
or Brian Armstrong, how active they will be in the
midterms in twenty twenty six and then ultimately in the
next presidential cycle in twenty twenty eight, always.
Speaker 3 (15:58):
Pushing us forward. Kaylee me, thank you.
Speaker 4 (16:01):
Now, let's talk about artificial super intelligence.
Speaker 3 (16:05):
Perhaps it's not far off.
Speaker 4 (16:06):
Last months off Bak CEO Masayoshi's son and predicted then
it would become a reality by twenty thirty five. Well,
former Google CEO Eric Schmidt, who is out with a
new book on artificial intelligence, says it might be coming
even sooner than we think.
Speaker 3 (16:19):
Take a listen.
Speaker 9 (16:20):
It's probable that we can build systems that are the
technical term is super intelligent, where you have a simple
system sarry a single system that is at the nineteh
percentile of physics, math, chemistry, and arts and so forth.
Speaker 7 (16:36):
No human can do that.
Speaker 9 (16:38):
It looks like these systems will be not only available
in the next five years because we already have examples
of passing these tests already, but also that they'll be
broadly available for all society.
Speaker 4 (16:49):
A bit more on AI app because happening today in
New York the Evident AI Symposium, where some of the
most senior leaders of the world's banking and artificial intelligence
areas came together. They're discussing the AI adoption in the
banking sector based on the latest data from an AI index.
It's a global standard benchmark for aimaturity in banking. I
sat down with JP Morgan Chase chief Data and Analytics
(17:11):
Officer Teresa heitsen Rada and asked if look, the efficiencies
from AI are the number one return on investment.
Speaker 3 (17:19):
Take listen.
Speaker 1 (17:21):
The magic question is what's what's the return?
Speaker 10 (17:23):
Right?
Speaker 1 (17:24):
And the answer is very, very difficult to quantify, And
I think at this stage of the game, the return
is a little bit of efficiency for a lot of
people every day, but not something that you can actually
hang a number on. So I think that the next
generation of where we see this heading is, rather than
just using it to help you write your email or
(17:44):
summarize a document, is when we start to plug the
technology into knowledge bases, when we start to actually look
at the workflow of a particular person or group and
linked together the steps in that workflow with the tools
that they need to be able to do their So
you basically go from the five minutes of efficiency to
the five hours of efficiency. That's a road like that's
(18:07):
going to take some time to actually get there, but
I think that's where we see it evolving. So there
are near term benefits, but I think the most compelling
is just putting the technology in people's hands and letting
them start to experiment and understand how it actually works
and how it can be used.
Speaker 4 (18:24):
JP Morgan Chase Chief Data and Analytics Officer Teresa heisen Weather.
There I mean, while coming up, it's official the DOJ
does want Google to divest its Chrome web browser.
Speaker 3 (18:34):
All details next.
Speaker 11 (18:34):
This is bring big technology.
Speaker 4 (18:50):
The Justice Department and a group of states do indeed
suggest that Google may have to divest its popular Chrome
web browser and imposed limits on its Android operating system
in order to address anti trust concerns, all confirming earlier
reports by Bloomberg.
Speaker 3 (19:04):
For Now, Bloombergsley and Nyland.
Speaker 4 (19:06):
Joins US and it was your reporting that ultimately has
been confirmed. Just tell us through how realistic still this
divestment would be.
Speaker 12 (19:16):
We'll see probably in early spring. The Justice Department has
said that they want Google to be forced to sell
off Chrome.
Speaker 3 (19:24):
It's not clear who would buy it.
Speaker 12 (19:25):
You know, the top two other browser makers, Microsoft and Apple,
who have the cash, the Justice Apartment probably wouldn't want
to let them, so the next you know, most obvious
candidates are probably people like open ai or Perplexity or
some of the other AI startups that would really love
to have this kind of an avenue to customers for
their AI search goals. Other things that the Justice Department
(19:51):
put in there is that they want Google to have
to license a lot of its data. That's probably a
more realistic thing that would have to happen, and and
it would definitely give a boost to other search engines
and possibly AI startups who would now.
Speaker 3 (20:05):
Not have to scrape the web themselves.
Speaker 4 (20:07):
Alphabet once again one of the key drags on the
NASTAC today. Investors worried about this alphabet itself, worried about
it coming out, and still once again trying to say
that this is a radical turn of events if imposed,
how complex is it to actually see a company sell
off such businesses? And how much do we have to
(20:27):
now wait for the legal maneuvering to play out? They
now have what a couple of months to respond.
Speaker 12 (20:33):
They have until next month to respond, and then the
Trump administration in March gets to have another reply. So
if the Trump administration wants to change this proposal a
little bit, they are allowed to. We probably won't have
a decision from the court until next summer.
Speaker 7 (20:48):
The judge has promised.
Speaker 12 (20:49):
That he will rule by August twenty twenty five, but
then Google says it will appeal. That'll take another year,
So at the earliest we're looking at them having to
do this in probably twenty twenty six. But I think
people are recognizing that Chrome is actually pretty important to
Google's business. It is, you know, an avenue through which
a lot of people access it's a search engine, but
(21:12):
it also is how Google collects a lot of the
data that underlies its advertising. You know, whenever a user
logs in, it keeps a lot of information about the
browser history.
Speaker 3 (21:23):
You know where.
Speaker 12 (21:24):
People are going online, what they're looking at, and then
Google uses that data to help sell advertising. So this
could have knock on effects to how some of Google's
core businesses.
Speaker 3 (21:34):
Me and Eilan appreciate the update. Thank you.
Speaker 4 (21:37):
Now it's time for talking tech, and first up, we're
talking PDD chares as you see plunging after warning that
it's profitability will trend downwards over time because of intensifying
competition in its home market. Of course, of China, PDD,
which competes with Ali Barba and its team, is struggling
to catch up with unspecified rivals because of the lack
of expertise, speaking of which Ali Baba has a pointed
(22:00):
better and executive Jeng Fan to oversee its entire online
e commerce operation. Now Jiang will lead a newly created
department that consolidates Ali Baba's online shopping assets, making him
the most powerful person in the company after the CEO
and chairman and Baido. It recorded its biggest revenue drop
more than two years after China's economic malaise. Under mind
it's push into general to AI. Now the company's ernie
(22:21):
Bot has fallen behind Bite Dances d'albao in China usage,
while its core business is losing ground. Two newer social
platforms off by seven percents.
Speaker 3 (22:30):
You'll see now coming up.
Speaker 4 (22:31):
We're going back to the US earning story Nvidia the
impact on the chip sector at large. We're currently off
by one point three percent, one point two percent for
the world's most valuable publicly traded company. This is the
room Mate Technology. Welcome back to Blue meag Technology and
(22:57):
Karin Hider, New York. Let's get back to it. In
Video's earnings. Here is the reaction from just some of
our guest somebody Meg Television.
Speaker 3 (23:05):
I have a feeling that we've reached peak Navidio. How
long can you stay perfect? Great?
Speaker 13 (23:09):
So I think that, yeah, I mean, I do think
that a lot of investors are expecting a little bit
too much.
Speaker 10 (23:16):
There's a little bit of this kind of fatigue.
Speaker 7 (23:18):
But make a mistake. This was a remarkably good quarter.
Speaker 3 (23:22):
When they're putting up in terms of Hopper, in terms
of overall growth, it's exact what the what the bulls want.
Speaker 13 (23:29):
Never underestimate Wall Street's ability to miss the bigger picture.
This company has double the operating margin of any other
Meg seven outside of Microsoft. To nitpick in Video on
those margins is truly missing the big picture.
Speaker 3 (23:43):
Let's nitpick some more.
Speaker 4 (23:44):
Daniel Pelling, portfolio manager and senior technology research analyst at
Sans Capital joins us now and Daniel, Yeah, we're quibbling
on a margin of seventy five percent going to somewhere
to seventy three to seventy four percent, so much more
profitable than rivals.
Speaker 14 (24:00):
Yes, Hi, good morning. Yeah, so well, I think actually
the gross Martin stuff is largely behind us. They've guided
for sort of seventy five percent exiting the year, And
I know this was a concern a while back, but
I would actually argue it's somewhat frankly irrelevant. I think
what was much more important was two things. First of all,
(24:20):
the company noted that the scaling laws are well and
alive and that we can both scale training and inference
going forward. So this entire discussion about how we're hitting
a wall is sort of probably off, and there's just
many walls and they're far away. And then the second
really important thing I thought that they mentioned was is
to say that data centers are becoming a little bit
like smartphones, i e. They are electricity constrained. And what
(24:44):
that means is just amazing, right, because that means every
year the hyperscalers have to buy a new chip because
the new chip is better in terms of electricity usage,
and we've never had that before in the data center.
So those two things I think are very positive for
the medium and long term of the company.
Speaker 4 (24:59):
I want to go about to the scaling laws, Daniel,
because yes, we've had Jensen Wang actually back on the
conference trail just earlier this week talking about how compute
power is going to continue to extend, and indeed the
more data, the more compute you have, the more sophisticated.
Speaker 3 (25:12):
The larger language models will become.
Speaker 4 (25:14):
But then you listen to the reporting out of Bloomberg
around open ai and the latest iteration of large language models.
Just stuttering a little, Why are you so confident that
scaling laws.
Speaker 3 (25:23):
Aresketden in place?
Speaker 14 (25:25):
Yes, So, I think the key thing to note here
is that all that information that we have is based
off of smaller clusters. So now we for the first
time have a cluster that's something like one hundred or
two hundred thousand GPUs that is owned by Elon Musk
and we'll see what comes out of that in the
next twelve months. And secondly, and more importantly, maybe the
(25:47):
black clusters start at one hundred thousand GPUs. Hopper started
at ten thousand GPUs, so we'll have to wait the
next twelve months. What can we train with these much
better chips and much larger clusters and whatever out of that?
That is the crux. So I think most of the
stuff that's in these articles is kind of backward looking
because we just didn't have the chips nor these big
clusters yet.
Speaker 4 (26:08):
That has been reporting about the latest valuation of Xai
and the fact that they're raising money simply to buy
more GPUs, But there is that worry about density of
customer for n video. Are we going to see a
broadening out?
Speaker 14 (26:23):
Yeah, So I think the big question here is like this,
and I think you're right. So the big customers are hyperscalers,
venture finers, startups, and actually healthcare companies are using training
a lot too. My sense is that the most important
thing to look at here is for the next year
is agentic AI and this idea of that we're spending
much more compute on inference and letting inference think. And
(26:44):
I think it's starting just now, and if we can
see this, then companies can really start using this via
using service now, Workday, Microsoft, et cetera, et cetera. So
I think that might be a big tipping point and
we're literally at inning one of that. And just to
be clear, agentic AI is this sort of THISI they
can do things autonomously within an enterprise and help with workflows.
Speaker 4 (27:05):
We've had a lot about agents from the likes you
mentioned service now we've also heard it from Salesforce.
Speaker 3 (27:09):
And the light Daniel.
Speaker 4 (27:10):
So when we look at your funds managed what the
top holding is in Nvidia, does that start to broaden
out for you or do you mean comfortable with your
alignment on in video even though it's at what the
most valuable market capitalization in the entire world.
Speaker 7 (27:27):
Two things.
Speaker 14 (27:28):
One, so I think there's a good scenario to be
made that Nvidia becomes the first ten trillion dollar market company.
And I'll tell you why. I mean, even if you
just look at the top four hyperscalers, they're estimated to
generate about three hundred billion dollars of free cash flow
next year alone. And what I'm trying to say is
there's a lot of dollars left to go to Nvidia
(27:49):
if the scaling.
Speaker 7 (27:50):
Laws hold true.
Speaker 14 (27:51):
So that's one, we're very bullish on video or continue
to be positive on video. But two, the trends are
really proliferating into other spaces, and maybe I'll make two
comments there. First, First of all, the software company Service
now Microsoft et Centa, they will benefit. But there's also
other companies you know that have significant competitive advantages and
it can really take AI and monetize it right. And
one great example might be a company called Axon that
(28:14):
you know, manufacturers tasers and cameras for the police force.
They've added an ability to transcribe you know, police when
they're engaged with people, and that says forty percent of
time for that police officer. That's a huge improvement, and
the stock has started to price them in as an
AI winner, and I think you'll see more of that
over time too.
Speaker 4 (28:32):
Another company for us to focus in on for our
audience standing pelling. So great to have time with the
senior technology research analyst, Sam's Capital. How interesting that he
kept referencing inference. We're going to have a conversation on
that now because we have a perspective and someone in
the field. Rodrigo leac CEO of sa Manova is an
AI infrastructure company that delivers advanced AI solutions for enterprises.
You're kind of an Nvidia competitor upstart.
Speaker 3 (28:56):
And I just want to articulate for our audience what
it is that you do. Yeah, no, thanks for having us.
Speaker 15 (29:00):
So some of the is started as a startup coming
out of Stanford University and palto really focus on efficient computing.
And so as we're seeing in video scale and you
see all the all the great numbers that that people
are talking about with video or we're now seeing, is
what happens when the world wants to go into production
and going to scale. And one of the things that Franklin,
we're not talking enough about is the fact that power
(29:22):
and energy is going to be a big bottleneck, and
you know, people are talking about building nuclear power plants
and things like that. Somanova Te took the focus of
how do we actually take AI at scale power efficiency
and drive it down so that you can actually scale
without running into these limits that we start seeing with
both companies and countries running out of power and figuring
(29:44):
out how to actually provide enough energy to supply these GPUs.
Speaker 4 (29:49):
So your view is you can't like there's going to
be a certain limitation on the amount of energy efficiency
that a company like and Video is ever going to hit.
With the GPU whereas you of a lot.
Speaker 15 (30:01):
Well, so today, for example, you know, a single rack
of Nvidia is north of one hundred kilowatts, and just
to give your audience a sense, that's one hundred, it's
one hundred homes that it powers. And so if you
think about someone of technology, we're able to drive ten
next to performance at one tenth the power now because
of a completely different architecture. As you know with n video,
(30:22):
it's an architecture that's been around for several decades and
so we're trying to evolve this graphics GPU architecture into
AI enabled now for inferencing, there's so many new techniques
that you can actually apply because you don't need to
do all the heavy computing that training requires, and so
you're able to drive a much much lower footprint in
terms of power, in terms of costs, in terms of space,
(30:44):
which ultimately is required when you're actually going to scale.
Speaker 7 (30:48):
And my belief is AI.
Speaker 15 (30:50):
Inferencing is going to be ten times more of a
footprint than training, and so I think we're going to
have to think about how to be energy efficient, how
to be cost efficient before it gets to the scale
that everybody get.
Speaker 2 (31:00):
To have it.
Speaker 4 (31:01):
Can you for our audience that is not only an
investing audience, but really a technologically savvy audience. Describe to
us what it is that's different that enables this higher
energy efficiency within your particular.
Speaker 3 (31:13):
Type of technology, because it's not GPU at all.
Speaker 15 (31:15):
What is it, right, It's not graphics and so it's
something that was inventing researchers stand for University. It's called
data flow architecture, and what it does it allows you
to actually map the hardware in the way that a
neural net wants to operate. And so with n video today,
you're having to break up these neural nets into these
cores that exists inside the graphics chip, which turns out
to be a lot of work, a lot of energy,
(31:36):
a lot of power salmon over with the data flow
architecture allows you to avoid all of that extra power
that's be incurred when you actually break up these neural
nets and let it flow through the machine, and so
you're getting these order magnitude improvements in power efficiency, which
ultimately allows you to be able to deploy these in
existing data centers, which, by the way, is one of
the things that I think is not talked about. When
(31:58):
we actually acquire GPUs because you also have to upgrade
the data centers where for our technology going into existing
data centers with existing power grids, with existing cooling systems,
it becomes one of those things that we allows us
to generate value for enterprises really really quickly.
Speaker 3 (32:15):
What enterprises you already locked in with?
Speaker 4 (32:17):
How are you managing to sell and this narrative into
corporates and federal government?
Speaker 15 (32:22):
Yeah, we're yeah, we're as a startup again and videos
at the top of the tier. We're a startup that's
you know, coming in and then providing disruptive technology. We're
the most deployed in the US government today as a startup.
Speaker 3 (32:34):
You know, we're going to go into national abs.
Speaker 7 (32:36):
And places like that.
Speaker 15 (32:37):
And and today we power also Saudi Aramco's Metabrane, you know,
so they're internal GPT. This is a private GPT with
data trained into it and now operating air gap within
the company. And so companies that worried about their data
their secured their privacy can now train these models on
assemblent of platform completely securely and then deploy within their
(32:57):
own infrastructure because you don't have to expose the.
Speaker 3 (32:59):
Data outside your firewalls.
Speaker 4 (33:02):
There are other startups in this space, or indeed trying
to offer more energy efficient options and pay for example
also back by Oracle. You yourself used to work to Oracle.
I'm interested as to what your next step is to
really gain market share.
Speaker 3 (33:17):
Is it about raising funds?
Speaker 4 (33:18):
I know you've got big VC backing for example.
Speaker 7 (33:21):
Yeah, I think you know.
Speaker 3 (33:22):
It's really good point.
Speaker 15 (33:23):
My belief when we started the company back seven years.
Speaker 3 (33:26):
Ago is about access.
Speaker 15 (33:27):
When the marketized AI, you got to make it available
to everybody. As your previous guests just said, is today
in video is really in large scale only to a
few customers. What we want to do by collapsing the
power is to be able to make it available to everybody,
to make it easily available, and that means lower cost,
much lower power and then able to deploy in very
(33:51):
tight spaces and not regarding needing and giggle what data center.
So some of that we focus very much on completely
restructuring the power, the power requirements of the AI infrastructure
so they can deploy racks at performance that much higher
than in VideA, not less than VideA, but much higher
than nvidea and much lower power.
Speaker 4 (34:11):
Samonova Rodrigo Lean CEO talking about some of the alternatives
out there. John and video, We thank you for coming on. Meanwhile,
coming up Palo Alto Networks earning. They are looking at
the first quarter that is proving itself with this focus
on platformization. We'll dig into that word in a moment.
CEO Niqushaurra joins us. This is Blue Bag Technology, cybersecurity company,
(34:45):
Palo Alto Networks, releasing earnings after the market Wednesday. And look,
we're seeing revenue climbing fourteen percent, earnings climbing thirteen percent.
We're seeing a focus on platformization and also a stock split.
We're currently up one and a quarter percent. But let's
get to the details with the CEO Nikosh Aurora of
Palo Alto Networks without some volatile trading for the market
now pushing you higher. And I'm just interested on the
(35:07):
nu once you are trying to tell.
Speaker 3 (35:09):
A different story to your investors.
Speaker 4 (35:10):
You're talking about platformization, you're talking about bundling. But still
they want to turn to your billings number. How do
you manage to get that across to them that stopped
looking at it because there was a little bit of
worry about billings.
Speaker 2 (35:21):
Well, first of all, thank you for having me, Carol,
and I appreciate you saying the word platformization, because this
is something we've been trying.
Speaker 7 (35:27):
To tell the market.
Speaker 2 (35:28):
I think we are seeing a c shift in the
cybersecurity industry where people are tired of buying point products
which are not solving the problem. Today, you have to
make these products work together. And I was listening to
the conversations before this. Everybody's talking about AI. AI in
our business means things are going to happen faster. Attacks
are going to happen faster. You're going to have to
(35:48):
resolve them faster, which means you need a common language,
you need common data which actually points towards integration and
stuff that works together, or points towards platformization.
Speaker 4 (36:00):
Cloud platform strategy that seems to be earning you plenty
of companies.
Speaker 3 (36:04):
What is it accumulative?
Speaker 4 (36:05):
You've got one, one hundred platformizations?
Speaker 3 (36:09):
Ultimately, is that more valuable? Is that more reven generating?
Speaker 4 (36:12):
Because people are worried that maybe within this bundling there
is some discounting going on.
Speaker 2 (36:17):
Well, I think it's important to understand once you deploy
a platform and it works well for you, you're not going
to replace it. So You're coming from an industry where
you look, we've had companies that have worked well for
eight or ten years, but they've been through a cycle,
and then you have to replace them and replace them
on a net new set of vendors because the last
ones didn't stay innovative enough. And what we're trying to
do is we're trying to stay evergreen. We're trying to
(36:38):
be innovative. We're trying to make sure we embed the
platform as part of your security solution that you logically
come to us for the next piece of innovation. And
you have to go through a huge process of constantly
transforming your security. In state a state today, more customers
have security transformation projects on tap than any other technology
transformation because the security landscape is constantly changing.
Speaker 4 (37:00):
Talk about the cyber landscape and whether it changes with
a new administration. Is there going to be more consolidation,
and is there going to be more deal making from
you look, I get.
Speaker 2 (37:10):
Past that in different ways. You know, I'm watching it
like everybody else and saying it sounds like we're going
to see the new doge go out and do a
lot of cost cutting and deploying a lot of technology.
Now Traditionally, the federal government in the United States has
been sloughed off new technology rightfully so, perhaps because they
want to sense their way through it. But as you
see efficiency, as you see technology investment happening, it has
(37:33):
to be secure. So I think security gets a knock
on benefit if we actually see a technology wave coming
from an administrative perspective, Do I expect regulations to get easier?
Possibly from an M and A perspective, Do I expect
that things.
Speaker 7 (37:46):
Will move faster?
Speaker 3 (37:46):
Hopefully? So, I think generally it should.
Speaker 2 (37:49):
Sort of board well for all of us in technology,
But you know, we'll all find out.
Speaker 1 (37:55):
We'll wait and watch.
Speaker 4 (37:56):
Look as you innovate, as particularly within the world of
gender to AI, we managed to get more sophisticated cyber protection,
you're also getting more sophisticated hackers. The US government has
just flags and flaws with Palo Alto's owned products that might.
Speaker 3 (38:11):
Have been exploited by hackers.
Speaker 4 (38:12):
Can you just tell us a little bit about whether
that was a realistic one for.
Speaker 3 (38:16):
You, what patches have been made? Well, like what.
Speaker 2 (38:19):
Happens is because we're coming from a hardware industry where
we sell products, customers deploy them, there's a long product cycle,
and you've seen this across the industry. At any point
in time, you know, there's always some bug that is
out there that needs to be passed and fixed. So
we're very good about making sure we work with our customers.
We have full visibility into who has what product. We're
able to work with them to pass those things. I
(38:40):
think as we move from a hardware based solution business
to a software business, which the world is moving to,
that becomes a lot easier to fix. I think software
beats hardware from a security recency perspective all the time,
So as the world transforms there we'll see less than
less of that in the future.
Speaker 4 (38:57):
As the world transforms, everyone becomes an investor retail community
as well. Talk to us about your two for one
stock split. You're already up what thirty percent on the year.
Is that why you spread it out? Is it about
giving more access?
Speaker 2 (39:09):
Well, we want to make sure that there is ample
access to our stock to retailed investors. You know, five
years ago when I started, we were a sort of
a niche name from the cybersecurity industry. Or delighted that
we are now part of the general conversation. We want
to make sure everybody has access to our stock.
Speaker 4 (39:22):
Well for now, we'll currently see it's going to be
action in a two for one stock split. We'll see
how much adoption there is around that. But in cash
Aurora on the fundamentals and some of the share moves.
CEO of pallel to Networks, we appreciate your time today.
Speaker 3 (39:35):
Thank you.
Speaker 4 (39:36):
Coming up more earnings, Snowflake giving a pretty strong outlook
for its product sales growth.
Speaker 3 (39:40):
We dig in this is.
Speaker 11 (39:41):
Blue meg Technology.
Speaker 3 (39:55):
More earnings for you.
Speaker 4 (39:56):
Snowflake shares absolutely jumping after the company go some pretty
solid sales outlooks for us, suggesting newly launched products are
receiving a strong reception from customers. Brady Ford has the
inside take, I mean extraordinary move to the upside from Snowflake.
We know it's a volatile stop, but what's managing to
win over customers?
Speaker 10 (40:14):
Bernie in software right now, the big question is who
gets left.
Speaker 7 (40:19):
Behind in the age of Genai.
Speaker 10 (40:21):
For a long time, investors worried that Snowflake was focusing
on maybe a two narrow slice of the data market.
You know, the services used to kind of organize and
analyze all your data. And Snowflake is successfully making an
argument that no, in the era of Genai, when folks
need to understand such a wide array of data and
(40:42):
turn it into interesting applications, we will remain a platform
where they spend their money and.
Speaker 7 (40:48):
Make their decisions.
Speaker 10 (40:49):
And so that's what investors are reacting to in these results.
Speaker 4 (40:52):
They organize, they analyze your data, but when it comes
to unstructured data, they're having.
Speaker 3 (40:56):
To make purchases. Here, it's a bit of.
Speaker 7 (40:58):
M and a Yes.
Speaker 10 (41:00):
So many have heard of this famous rivalry between Snowflake
and Data Bricks for a long time. The argument was
that data Bricks handled the unstructured data better, and that's
the real you know, something that's not in a spreadsheet,
something that's real disorganized. Maybe it's a bunch of Google docs.
You want to be able to scan through and say, hey,
give me insights. And so Snowflake has worked to beef
(41:21):
up their ability to deal with that kind of hard
to access data. And yeah, last night we saw them
make an acquisition of a startup that focuses specifically on
ingesting this kind of messy, unstructured data you need for
a lot of Genai use cases.
Speaker 4 (41:37):
I know that you've reported a lot on that rivalry.
So see if that M and A does indeed help.
But under the guidance of the new leadership in Ramaswami,
he's also trying to integrate large language models, right, how
is this being seen in the anthropic deal Exactly?
Speaker 10 (41:52):
They want to bring more LLLM access right into the
Snowflake platform. So if you want to do things like
against stand through all your information and pull out insights,
make dashboards, make visualizations, you can do it right in Snowflake.
You don't have to go open up a new you know,
Azure instance or leave the platform. So bringing anthropic models
(42:14):
right into Snowflake is again hoping to make it a
platform where you remain where kind of your most important
data is. They really hope to be a center of
gravity for all your information and not just something you
you know, use when you want to make an analysis
of a spreadsheet somebody sent to you.
Speaker 4 (42:33):
Brady Ford, across the Snowflake moves and indeed some of
the intricacies and the earnings we thank you. Look, let's
just get a border perspective of how snowflakes thirty percent
the upside fits into what's happening more broadly across the
NASDAK because look, it hasn't been an round out risk
on kind of a day. In fact, we've been lower pulled,
lowered by about two tens a percent, large part because
(42:54):
of Nvideos off by one point seven percent, Alphabet two
on the back of that formalization of Bloomberg reporting. Then
indeed the DOJ does want it to be selling off
Chrome for example, but in video of by one point
seven percent is important because it is the most valuable company.
It is going to have a huge points perspective when
it drags lower.
Speaker 3 (43:10):
But look, we were.
Speaker 4 (43:11):
Anticipating in the options market there's mure an eight percent
swing upside or downside. So this is relatively stable, shall
we say. After earnings did show a ninety four percent
increase in revenue, a seventy percent increase in revenue for
the Guide, and margins being pretty strong.
Speaker 3 (43:26):
Interestingly, the chops that the.
Speaker 4 (43:28):
Socks the chip sector is up some seven percent, so
the rest of the market doing well today. Have a
look at what's happening with Alphabet because the parent company
of Google off another six point eight percent today. Once again,
this anxiety surrounding antitrust and how it might have to
be split off weighing on the stock that does it.
In this edition of Bloomberg Technology, a lot to get through.
(43:49):
You want to check it all out again on the podcast.
You'll find it on the terminal, as well as online.
Speaker 3 (43:53):
On Apple, Spotify, and iHeart. This is Bloomberg
Speaker 7 (44:00):
Today, The Healthy Everyday, Healthy Everyday, The health of
Popular Podcasts
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
The latest news in 4 minutes updated every hour, every day.
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.