From the heart of where innovation,
money and power collide in Silicon Valley and beyond.
This is Bloomberg Technology with Caroline Hyde and Ed Ludlow. Live from San Francisco.
This is a special edition of Bloomberg Technology live from the Bloomberg Tech
Summit. Yes, absolutely.
Leaders from across tech and business are gathered here to discuss how they’re
going to navigate economic geopolitical uncertainty and multiply the
opportunities in that landscape. We’ll be hearing from some of those
speakers. We kicked it off last night with
incredible conversations with Alphabet’s Sundar Pichai, Messers Andrew Bosworth.
Right now, the tech sector is also very focused and markets are right about the
discussion between President Trump and China’s Xi Jinping, and it’s giving a
positive spin to markets. Nasdaq coming out of its lows.
We’re now trading higher. Look, not gargantuan moves.
We’re up about three or 4/10 of a percent on the Nasdaq 100, as you see.
But we are trending higher as we see a positive signal coming from President
Trump that they’re going to be meeting yet further with, of course, the first
lady going to China and indeed a reciprocal kind of discussion coming
with the Chinese leaders, coming to us from more broadly rare earth.
Is there any clarity on where we want to have?
Let’s get some news breaking out of the Bloomberg Tech summit and drill the
defense tech company that’s quickly become a major player in US National
security has just closed its latest funding round.
Joining us for more is Tracy Evans, co-founder and executive chairman of
Pandora. He’s also a partner at Founders Fund.
He will be a speaker as well at the tech summit later today.
Tracy, good morning. Good morning.
Congratulations. This is series G.
What are the details? And this seems meaningful for andrew.
It is. You know, as we continue working on
building a company that has the capacity to scale into the largest problems for
the national security community, we thought it was really important to shore
up the balance sheet and make sure that we have the ability to deploy capital
into these manufacturing and production problems that’s that we’re working on.
The round ended up being a two and a half billion dollar round at a valuation
of 30.5 billion post-money It’s a tremendous opportunity for us to
continue growing the team, growing the infrastructure and our ability to really
ramp up into these into these bigger problems.
That sound is fund led the rounds. You have an interesting role as
executive chair at and Daryl and I but you recuse yourself from the Founders
fund part of that I believe. Right like one of one of if not their
biggest check ever. Yeah the largest check in Founders Fund
history. We put $1 billion into the two and a
half billion dollar round. Okay.
Why is this commitment, this necessary money right here, right now, Is this
going to be the last one before potential IPO?
Oh, man, it’s a great question. You know, there are a lot of ways to
raise capital. It’s not just the possibility of IPO.
Obviously, long term. We continue to believe that Andrew, is
the shape of a publicly traded company. We’re not in any, you know, rapid path
to doing that. We’re certainly, you know, going through
the processes that are required to prepare for doing something like that in
the medium term. But right now we’re just very focused on
the mission at hand, going at this as hard as we can.
Does that mean we raised another private round of capital at some point?
Maybe Hard to say. You know, we’re always surprised that
how much demand there is in these massively oversubscribed and how
oversubscribed based it was a lot like 8 to 10 ex of what we had the capacity to
take on. You know there’s a lot of players that
are starting to figure out that the momentum is drawing into you know being
able to do this at Andrew. And I think there’s just a lot of a lot
of interest from growth funds for crossover funds, for mutual funds, for
getting a bite of that before there’s a public market opportunity.
Well, interesting you say getting a bite of the action when it comes more broadly
to defense, suddenly want to Silicon Valley kind of wants to be in defense
and in US manufacturing. In fact boss Andrew Bosworth of matter
yesterday talking about how maybe there’s some bigger patriotism than we
realize across Silicon Valley. Just take a listen.
There’s a much stronger patriotic underpinning that I think people give
Silicon Valley credit for. But taking a step deeper, this is this
is actually a return to grace for the valley, potentially not just from Meta,
but from Google and other companies. The Valley was founded on a three way
investment between the military, academics and private industry.
That was the founding of it. There would be no technology if we
weren’t all tasked with the problem of computing naval ballistic trajectories
during the first two World Wars. You have a partnership.
The technology the metal makes working hand in hand with Andrew.
Interesting. Of course, that malarkey now goes back
to a company that he of course sold Oculus to in many ways with this
partnership. What does it look like in more
partnerships going to be eroding, erupting here?
Well, I mean, to to Bob’s point, you know, the world has changed a lot in the
last in the last eight years. Andrew actually turns eight years old.
Tomorrow, we’re approaching our anniversary, which is the anniversary of
D-Day as well. We started the company on the 73rd
anniversary of D-Day, recognizing there’s an important contribution for
not only ourselves but also the American enterprise into national security
efforts. You know, Palmer was fired by Matta in
2016 for his politics. You know, those Google Maven walkouts
and protests. The reality is, is that that the silent
majority actually believes that these things are really, really important to
work on. And, you know, polymer being able to go
back to its roots and reach a point of forgiveness with the media team to focus
on the most important task at hand, which is leaning in on the best in class
optics and VR work that they’re doing to contribute to the mission that we’re
leading with, the IBEX opportunity with the DOD.
What was really interesting about how Buzz explained it is that they are
basically providing components. We actually had DeLeon asked for off on
the show the other day in the context of Founders and Enduro sat.
And what he was talking about is that the innovation of big tech is being
driven by consumer electronics. Right.
And thinking about your supply chain. Could you just explain the mechanics of
how metal works with you and why it’s useful to you for them to come up with
these components in the in the products that you want to do?
Right. I mean, they’re obviously they’ve done
an incredible amount for the AR VR community.
I think Palmer has been on record many times saying that Zuckerberg is VR boy
number one. He’s led the way with putting a ton of
capital into solving some of these basic science and research problems around
optics. You know, the physics challenges that
make this really difficult. You know, we don’t have to take all of
that on ourselves. We can go and partner with someone in
America who believes in the mission, who’s done a lot of that work, and then
we can bring that into an integrated system, which is really what we do as a
defense products company. Let’s talk politics.
You talked about Palmer’s politics. You know, the story today is about
President Trump speaking with China’s Xi Jinping.
And at the heart of that is rare earths. But I actually think for you guys,
you’ve historically, when we’ve discussed China taking a different
approach, which is you believe and role is critically important now because of
future hostilities from China. Just explain that, that thesis.
Yeah, I mean, obviously there’s a huge supply chain component to all of this.
China is right in the middle of most of these conversations around manufacturing
production, supply chain challenges with natural resources, including rare earths
metals. You know, at Android, we’ve been
thinking about this long before the you know, the second Trump administration
started. We’ve been shoring up to make sure that
our Tier one suppliers are being sourced from allied nations, the US and our
allied nations. And, you know, for the first time ever,
I think that the entirety of the government is taking supply chain and
seriously as a national security asset rather than just something that’s kind
of like important from a trade context. So I think it’s incredibly critical that
we don’t wait until we have, you know, our access to critical materials, our
access to chips, our access to all of these things removed from us.
But we need to start happening. Isn’t there a limitation already on rare
earths? And how does that impact your business?
There is and you know, this is the type of thing that you have to plan for well
ahead of time, believing that at some point, yeah, we might lose access.
And I think all, you know, intelligent and thoughtful American companies need
to be thinking about how to how to address that problem before it actually
becomes a really critical existential threat.
What has the administration been doing it in a methodical manner that you
wanted to see? Because at times there’s been a worry
about basically the timeframes that we get cut off too quickly before the US is
really understood its own supply chain and resources.
Many companies haven’t been looking at this as well as Andrew.
Absolutely, yeah. I mean, at Enderle, certainly we’ve been
thinking about it a lot and through the Founders Fund context, I spend most of
my day thinking about this exact problem.
You know, G has said very clearly that he intends to reclaim Taiwan by 2027.
That’s a really short timeline. It doesn’t mean that we have five or ten
years to build up semiconductor manufacturing capabilities.
It means that we have like two years, 18 months.
And so we really need to be thinking about these things.
I think it’s incumbent upon the US tech community to contribute in ways it’s
incumbent upon the US venture community to put capital towards solving some of
these problems. And certainly every company needs to
have a plan for how they’re going to address that should it become a real
issue. Golden dome.
Going down? Yes.
You know, obviously, there’s very few things as exciting as the idea of
protecting the US homeland from from airborne threats.
You know, we’ve been engaged in a lot of conversations there, too.
Too soon to comment on anything. That’s my understanding, though, said
when when this big announcement happened at the White House, there were some of
the sort of legacy prime defense contractors named Lockheed Martin being
one. And a lot of people that I speak to were
surprised that the name and Israel wasn’t said over and over again because
of the position you now have as a supplier to the US military apparatus.
But it is early in this process. I guess the point is what will happen?
Do you have to make pitches and competitive bids and say, here’s what we
can offer in this? Or is it a case of the administration
coming to you and say, do this for us? Well, it’ll obviously have to be a
tremendously multi-layered approach. There’s not going to be a single vendor
that provides the entirety of the system.
You know, there’s in the factors, interceptors, things like that, space
based interceptors, you know, all the sensor networks that are going to be
required and then a software layer to coordinate and see to all of that
activity that’s going on. So, you know, it’s not just going to be
an arrow. It’s not just going to be Lockheed
Martin. I believe that it will likely be a
mosaic of a lot of different companies coming together to make sure that this
problem is solved in the best way possible.
You talk about space. There is one particular founder who is
very close to the White House who now seems to be less close and is Elon Musk.
How are you seeing the relationship of venture with the administration right
now? And is there a repercussion as Elon
withdraws and then seemingly gets pretty angry at the state of the current
spending bill, or at least the way he sees it, as a spending bill, as
reconciliation? Yeah.
Well, I mean, we’re still very kind of near in on the start of the
administration. We’re only, what, five months into
things right now. I do think that this administration is
incredibly communicative with the tech community, with the venture community.
The channels are open. They’re curious.
They’re looking for for information. We have a direct line with President
Trump. There’s lots of paths into the White
House. Without without without commenting any
further on that. But yeah, I think I think that there’s a
tremendous openness to having that dialogue.
A lot of these initiatives, whether you’re talking about the big, beautiful
bill, you know, I like what Elon said. It’s difficult for a bill to be both big
and beautiful. It can be big or it can be beautiful.
It can be both. But, you know, I think, you know,
there’s there’s at least receptivity.
And we’ll see how some of these big programs, these big plans shape over the
next year. And the reason Carrie’s question is
important is, you know, you are part of the contingent that went to the Gulf
War, the Middle East with the president and corporate America.
Let’s end here with a look at the future.
After you and I were together in Ohio in January, my phone just kept ringing from
private companies and startups that are in the defense technology space.
Is there a world where Antrel and three or four other players exist and coexist?
Or is it just and a role and nothing else?
Well, first, I want to say thanks for coming to the Arsenal launch and it’s
freezing cold. Thank you for using gold.
You’re one of the few people Xandros who has even seen the site in Ohio.
So it was cool to have you out. Thanks so much for doing that.
Yeah, you know, I think categories are tough, right?
Like, there usually isn’t such a thing as category investing.
You know, if you’re a social media investor and you didn’t invest in
Facebook, you probably lost money. If you’re a space investor and you
didn’t invest in space X, you probably lost money.
Do I think that there’s room for a couple of players to make a big entry
into the defense space? Sure.
I don’t think there’s going to be multiple Andros.
And that’s really our goal is we’re trying to create a concentrated
opportunity to be a next generation defense products company that goes out
and wins these major programs where we’re going to have the ability to bring
software chops, the autonomy, speed, scale manufacturing into the defense
industry for the first time as a new player in a really long time, a new
player that is now worth more than $30 billion, just double the valuation,
having raised more than 2 billion trace here.
And it’s a joy to have you and Doyle, executive chairman, Founders Fund
partner as well. Now, so much more coming up about
clearly the future of A.I. Alphabet CEO Sundar Pichai following his
panel avenue big tech summit last night talking about actually competition.
How is looking at A.I.? This is pretty high technology.
You know, there’s no doubt to me that three years from now there will be a
company which will be dominant in this age, which we don’t even know the name
of today. Welcome back to a special edition of
Bloomberg Technology live from the Bloomberg Tech summit.
Alphabet CEO Sundar Pichai says the company will continue to expand its
engineering ranks at least into 2026, emphasizing human talent remains key
despite air investments. That was part of the discussion with
Bloomberg’s Emily Chang last night when we kicked off the tech summit here in
San Francisco. Here’s some more of that conversation.
In 2025, our CapEx is $75 billion, right?
So. So we are definitely investing for the
long run. But, you know, it’s the same investment
which powers businesses from search to YouTube to cloud to workspace to Android
and play to Waymo. Right.
And so and to our subscriptions business, we just launched
Google I Pro and Ultra and you know, it’s definitely doing well I’ve been
pleased to see the reception. So I think it’s such a profound
technology. I feel the opportunity ahead is bigger
than the opportunity we had in the past. So to double click there, if I may.
What specific new revenue streams do you see increasing and what revenue streams
do you see decreasing the growth in cloud?
Like Vertex, say AI is up 40 X in usage on a token basis just in the last
12 months. So obviously, you know, we have billions
of dollars in and providing AI based solutions in the air and infrastructure
air subscriptions. So there are many, many new businesses.
You have over 180,000 employees right now.
Is it half that in the future? I expect we will grow from our current
engineering base even into next year, right, because it allows us to do more.
I think I think the opportunity space is also increasing.
I just view this as making engineers dramatically more productive,
getting a lot of the mundane aspects out of what they do, allowing them to spend
on higher value added tasks. But that means it’s an accelerator.
People will be able to do more, which means maybe will create new products and
hence we will need more people, at least in the near term.
To me, it looks like, you know, we will expand engineering
velocity and that doesn’t mean we are constrained in what we will do.
We’ll end up doing more as a company as well.
But in two trials now, judges have said that Google is a monopoly in search and
partly a monopoly in ads. How do you address the concern that your
AI is built on an existing domination of search and ads and that this is just
reinforcing original monopolies? First of all, you know, we disagree with
the rulings and we’re in the process of appealing these things.
Look, if anything, this moment has shown I don’t think there’s anyone here who is
using anything they don’t want to use. If you look at the success of Jeopardy
or any of the product and people are literally have more choice than ever
before, the reason people use Google is because they want to use it.
Right. And so I think I think we continue to
innovate. I think choice choice is good for users.
Competition is good for the world. So that’s how I see it.
You’ve said the remedies proposed are too extreme.
Would you ever voluntarily break yourself up, like control your own
destiny compared to what the initial scope of the ruling was?
Some of the proposed solutions are far overreaching.
We’ll see how it plays out. I’ve heard you say a few times that you
think of AI as an expansionary moment, but so far it does seem to be favoring
tech giants and well-funded startups with access to GPUs and data centers and
enormous amounts of capital. Like isn’t this really concentrating
power in fewer hands? And is A.I.
just another winner take all game? I’m confident that is a company that’s
going to be created with the AI, just like when the internet happened many
years after the incident happened, Google didn’t exist.
So, you know, there’s no doubt to me that three years from now there will be
a company which will be dominant in this age, which we don’t even know the name
of today. Yeah, that’s the only way things work in
the future. A confident alphabet CEO, Sundar Pichai
that along with Bloomberg’s Emily Chang. Right now we can talk it all through
with executive editor Tom Giles, who joins us for more that conversation.
You were sitting down with balls from Mehta after that with Cinda.
But just going back to what I was saying around his optimism on jobs today, one
of the most read stories on the Bloomberg terminal is Vista Equity
Partners Robert Smith saying, Look at my conference today.
60% of you are going to need to find new jobs.
White collar work is going to number keeps going up.
And I will tell you just last week, Dario from Jay from Anthropic 50%.
Right. Look, the fact is AI is definitely
coming for some engineering jobs. The number, the percentage of coding
that is happening from artificial intelligence seems to keep creeping up
and up and up. Some people would argue that’s going to
put a lot of engineers out of jobs. But as Sundar pointed out last night, he
thinks what’s going to happen is it’s going to make them more efficient.
It’s going to mean creating more products, bigger companies, bigger
businesses and need for more people. Where do you where do you draw that
line? I mean, we we have to encounter as
journalists, right? Right.
When is it and when when are you going to hear the news from the eyes?
We have to stay a step ahead of them. We have to stay smarter.
And the engineers, it’s the same thing. You have to become more sophisticated in
the way you code. And that’s what Sundar is talking.
And I guess the news headline was that they will keep hiring, at least for the
next 18 months or so. 2026 He didn’t want to be pinned down
into much longer into the future. The thing is that Google is so core to
the news cycle at the moment. It was that was why it was such a
important way to kick off pulling back tech summit.
He made this point that like, look what we’re doing in our scale.
You can’t accuse us of being a threat to the world existentially because we are
also facing all of this competition. Some of that competition is going to be
speaking over the next 12 hours, for example.
Absolutely. You can’t say on one hand that we are at
extinction level threat, right at the same time, too big.
He definitely he he came back to the antitrust question.
Right. The the basically the government right
now is considering breaking up Google. Would he consider breaking up himself?
Of course, his answer is that is not on the table.
He thinks that the government is overreaching right now and he has to
keep sending that message that we are not a threat and that there is a lot of
choice. Now, I will say that when you come to
him and say that people are using AI much more chat bots to do their
searches, even Tik Tok to do searches. That actually feeds into his argument.
That’s really beneficial to him to know that there is this much competition,
that chat, the chat is out there and people are starting to use it for all
the things that Google is, is you usually do with Google and perplexity.
Who’s speaking to you in a little bit later today?
Shery Ahn Ghaffari. We’ve also got Shirin speaking with the
government as well. Michael Kratsios.
What are some of the conversations going to really be
pushing forward, do you think? Because yesterday you got the matter to
push forward on this patriotism, focus on this alignment of Silicon Valley with
defense? What else came out of that conversation?
Right. Yeah, I mean, he pointed out that for
many years, Silicon Valley has said no to working with the Defense Department.
The pendulum is swinging dramatically. Patriotism, silent majority.
We heard that before in US history, you’ll recall.
Not sure I’ve seen the polling that shows that demonstrator, but I get the
sentiment behind it. As you talk just just now with Trey,
I’ll be talking with him a little bit later.
I’m looking forward to hearing it on. You.
Got to go. It looks great.
I am looking forward to hearing from Lurie, Mayor Lurie, to hear about the
comeback of San Francisco. Is it real?
Is it lasting? And what’s ahead for this important city
to Silicon Valley? We’re just getting kicked off here.
That was Bloomberg Technology executive editor, senior executive editor Tom
Giles. Let’s get out to Washington, D.C.
President Donald Trump and Chinese leader Xi Jinping agreeing to further
trade talks during a phone call today. Ramesh Kailey Leinz has all the details.
And at the center of it, rare earths. Yeah, that’s right.
The president said that after this phone call with Xi, which lasted about an hour
and a half, there should be no more questions around the complexity of the
rare earths issue. That does still leave us, though, with
some questions, as the president did not specify exactly what was agreed to on
rare earths. If China has formally said that it will
be lifting export restrictions or granting licenses for the export of
those rare earths to American companies. So we’re still left with some questions
here. China’s readout of this call has been a
bit different. China, in its statement, also said that
President Trump has said that Chinese students are welcome in the US, despite
the State Department’s efforts to be revoking some of those visas of Chinese
students. But the big picture here is that
President Trump says that this has brought the US and China to a better
place on trade and that they are going to continue conversations which will be
a bit broader. Remembering the first conversations in
Geneva where just the Treasury secretary, Scott Bessant, with his
counterpart now Howard Latin at the Commerce secretary, as well as the U.S.
Trade Representative Jamison Greer, will be involved in the next talks at a
location that has yet to be announced, as well as the time with
China. Top of mind.
Kelly in 30 seconds, Elon Musk also top of mind for this big bill.
Oh, absolutely. The speaker of the House was actually
planning on talking to Elon Musk today as he says he needs.
He wants to see Americans kill this bill, possibly in part because it will
roll back the 70 $500 EV credit by the end of this year.
And Tesla, of course, benefits from that.
Caroline. Katy lines me.
Thank you. Coming up, so much more from here in San
Francisco, Agility Robotics CEO Peggy Johnson on the latest in the robotics
space. This is Bloomberg Technology. Welcome back to a very special edition
of Bloomberg Technology. We are live on the big tech summit right
here in San Francisco. I’m Caroline Hyde.
I’m Ed Ludlow. Mr.
CTO Andrew Bosworth says that, quote, The tides have turned in Silicon Valley,
making it more palatable to the tech industry to support the US military’s
efforts. He sat down to discuss that last night
with Bloomberg senior executive editor Tom Giles.
I have family members in the armed services.
I want them to have the best equipment available and I have always been a very
patriotic American and believe in the importance of our role, not just inside
the nation, but globally. So there’s a much stronger patriotic
underpinning that I think people give Silicon Valley credit for.
But taking a step deeper, this is this is actually a return to grace for the
valley, potentially not just from media, but from Google and other companies.
The Valley was founded on a three way investment between the military,
academics and private industry that was the founding of it.
There would be no technology if we weren’t all tasked with the problem of
computing naval ballistic trajectories during the first two.
World wars like that is the heart and soul of the investment that led to what
we are today. And that really got severed for a while
there. And some of that is because of the
procurement laws that changed in the sixties and seventies.
Some of that’s the Last Supper that happened in 1985.
We ended up just kind of going our separate ways and I think we’re worse
off for it. I think we’re spending more money for
worse technology in the in the public sector.
And I think the private sector is not attacking some of the most valuable
challenges for us to solve. And I think it’s a mismatch.
You mentioned smartphones earlier. Do you see spectacles, headsets as a
replacement for or a complement to smartphones?
We’ve always we’ve always talked about the the mixed reality devices, VR
devices as being kind of a laptop style market.
It’s like I use this joke a lot, but, you know, if you’re ever on your phone
and you’re answering an email and you’re like, No, this is not a phone email,
this is a laptop email, and you’re like, Put your phone down.
You’ve got to upgrade your power. Like, Yes, now I have all the windows
and I will have all my research and I think there’s a laptop email, there’s
like an increase in power that comes there that we all recognize and they
just like know instinctively when you’re in my zone, the headset should feel like
that. It should feel like a crazy upgrade.
Like this is not a laptop email, this is a headset email just now, it’s like the
world is your oyster and you’re just moving things to the universe.
So I think there’s a there’s a story there certainly in the limit.
I think air glasses replace smartphones. That’s a ways off.
Smartphones are incredible. And it’s not just that they’re great
devices and they’re convenient. We’re used to them.
They also have like an incredibly entangled ecosystem of software
connected to the rest of the world around us that is going to be very slow
to move over. And a lot of people who aren’t going to
want to move over because it’s going to disenfranchise them or disintermediate
them. So I think I think that will take a
longer journey. The good news is they work really well
in concert. The platforms don’t make it easy.
Sinner does better than Tim. But the truth is, with air, you’ve got
this obvious use case, so it just makes sense to have these wearable devices,
even though you also still have a phone. So I don’t we’re not in a place anymore
where it’s like one or the other and you have to pick today.
They will coexist for a long time. Metis CTO Andrew Bosworth, along with
Bloomberg’s Tom Giles. Look, it’s not all about al VR glasses.
When you think about the future use, I think robotics for the perfect person.
Discussing with Peggy Johnson, CEO of Agility Robotics.
Now we are thinking about the next step of humanoids working with humans.
You already after a year in the job, a year with your humanoid robots in
logistics, what are you learning about how difficult it is for us to work
alongside them? Well, the biggest factor to scaling
humanoids is safety. So still today, humanoid need to work
inside of a work cell, which is a small gate that keeps them out of the
proximity of humans. To really scale.
You need to come outside of the work cell to be able to walk down to the
loading dock or across a factory floor. And we’re working on that right now.
Our next gen, which we will show off before the end of this year, will be
what’s called cooperatively safe. We can be within proximity of humans in
a safe manner. How much is that software dependent
rather than just hardware dependent? It’s a bit of both.
You have to have a lot of sensors, a lot of cameras.
Interestingly, unlike humans, robots can see behind them they can see three, six,
eight. They’re much like electronic vehicles in
that fashion. But you do need a layer of software on
top to ensure that nothing ever happens that would put a human in harm’s way.
Have you what range of tasks can your robots currently perform?
I guess autonomously, Literally movement pressures, Yes.
So right now that entry point is moving materials.
So moving fast. Two bins and boxes and things like that
in warehouses. But that is only the entry point.
It is a multipurpose robot. It can do a number of tasks, but this
entry point can teach us new skills as we learn to do more things in those
facilities and eventually come outside of those facilities and into our homes.
What is agility is core competence. You make hardware or are you better at
the software component? Well, I would say a little bit of both.
But you said that already. Yeah.
I get told every single day. At the moment, when you speak to a
robotics company, ask them about how good their software team is, how good
the engineers are, because it has been the limiting factor to progress.
Yeah, so A.I. has really supercharged that.
We used to have to code all the movements of the robot, and now with
A.I., we can teach it new skills much more quickly.
We’re actually changing, exchanging engineering hours for compute hours, and
it really shrinks the time and the cost of teaching the robot new skills.
What about supply chain? We’ve got a lot of news about China US
relationship today. How is it looking for you?
Well, a couple of things. We manufacture right here in the United
States, only about 1% of our parts come from China.
So we’re relatively immune to any of the issues that others may be having.
The robot was just initially built that way, and it’s been a real advantage for
us. Can you pit us?
Therefore, we love to put generative AI prowess of China versus generative AI
price in the US. What about robotics is winning?
I would say the US is winning. A lot of what you see coming out of
China today are flashy videos, demos not actually on a factory floor like we are
working eight hour shifts lifting heavy items.
That is well, that’s the true test of what, a robot.
That’s the thing. I think the Bloomberg tax audience
really just appreciate understanding the basics.
So with the GSO deal in one calendar year, you’ve moved 300,000 items.
But how many robots is that? How many actions can a robot doing?
A chef I can’t believe use the word shift performing know it’s just how does
it work in practice and give us a sense of scale.
Sure. So it is an eight hour shift.
It’s just a handful of robots that are producing that type of number.
And basically the robots can walk in and perform a task.
But what’s interesting about humanoids is that they can perform several tasks
so they can do one thing in the morning, something else at noon, something else
in the evening where robots, the traditional robots and say a single arm
is very specific. They do one thing over and over and they
do it very well. Humanoids can actually change what they
do. And now with A.I., we can give them
commands and they can change on a moment’s notice.
The history of agility. So interesting.
It came out of Oregon State University. We’re thinking at the moment about state
funding, about funding to science and R&D of talent being able to come in from
immigration. These things compromises.
Well, we did have early grants from the US and it really helped the company get
going. So I am a huge fan of basic research for
US companies. We absolutely have to continue to invest
both in research and in the talent. So we need proper immigration in place
in order for us to be able to access this kind of talent.
Peggy Johnson, CEO of Agility Robotics. And we’ll get a little bit more of a
demo later on in the day here in S.F.. Now coming up, Microsoft chief product
officer Aparna. Jennifer joins us next.
This is Bloomberg Technology. Welcome back to a special edition of
Bloomberg Technology Bill live from the Bloomberg Tech summit in San Francisco,
where AI is top of mind, AI’s impact on the workforce is already being felt, and
that’s especially true for engineers. Aparna Jennifer Gordon is the chief
product officer of Experiences and Devices at Microsoft, which is at the
forefront of some of these changes. She joins us here at Tech Summit.
So as you know, about six months ago, I just did 30 days straight of using
copilot on a Windows based laptop. I wanted to commit to like working out
how on earth do you use these tools in the real world?
Really interesting that I felt within the confines of the laptop.
I couldn’t get out of the laptop onto my smartphone.
What are you finding six months on and how knowledge work is?
Use your use your tech. First of all, thank you for having me
and also for thank you for like nailing the pronunciation of my name,
you know, like taking a step back. I think we are in this like super early
phase of like the AI shift. This is like the 1995 equivalent of the
Internet shift to me when I see it. And to me, I think what we are starting
to see is for knowledge work very clearly.
Efficiency gains, right? For ad organizations, you’re saving
millions of dollars of like money and time automating workflows.
I mean, summarizing documents, meeting notes, etc..
Is co-pilot just narrowly a tool for knowledge workers, or are you finding
consumers are using it in equal measure? It’s it’s it’s broad because
productivity is not bound by the work or organizational boundaries.
But I do think that actually one of the things that we are seeing where the
traction is, is in efficiency for organizations, but efficiency for at the
individual level to imagine having a digital assistant at work for every
employee. Right.
Which you couldn’t afford to have before.
The thing that we are seeing and I’m seeing it with customers talking about
it too, is that that shift between from efficiency to more of effectiveness,
meaning, look, I mean, it’s fine, it is great.
For example, I can’t imagine doing a meeting without the facilitation by
copilot. Right.
It takes the notes, It carries action items.
It tells if you’re going off the rails in a discussion, because in meetings,
that happens all the time. But what I am seeing is that for
effectiveness, the key unlock has been these model advances in deep reasoning,
meaning it’s not just about like faster answers.
It’s we launched an agent in co-pilot called Researcher.
I think about it as a as a super smart, intelligent chief of staff in your
pocket. And so, for example, I had a customer
meeting last week and this thing went around and kind of like scoured the
whole web information, but also the email from three years ago from the CIO
of the customer that they’re kind of like pointing and issuance and make sure
you bring that up. I mean, that’s the effectiveness.
So that’s like superpowers that you didn’t have before, that people are
getting their superpowers from other places too.
And your enterprise, the interesting nature of your relationship with open
A.I., but also them as a competitor, can your customers afford both?
Are they affording both? I think with that, with open air, first
of all, we have a great long term relationship.
It’s a win win relationship. We build products on innovative models
from open air. If the frontier models, our customers
build agents and, you know, workflows on and air ops on top of our Azure opening
service platform. What I do see at Microsoft is for us at
least what I and copilot is about, number one, bringing the advanced
intelligence, right? That’s table stakes there.
But number two is making sure that it works where you work.
Right? Right.
You’re in I’m in teams all day or I mean, not just where you work.
So right now I’m using this thing. This is a smartphone.
Various brands are available in the future because this is part of your
portfolio. Which device does Microsoft bet that
knowledge workers are actually using? Fatos.
I mean, there’s clearly a lot of knowledge work that happens on laptops
and desktops. But then I think we have a mobile app,
for example, for the M365 copilot, the one that you use at work.
But I think the other thing I do want to come back to is there is a you can’t
build all of these on Jenga blocks, right?
You have to build this on security, safety, compliance and governance.
And that’s the thing I hear all the time from customers.
In fact, the phrase that I use internally is we need to innovate at the
speed of trust. Right.
Which means it’s an end. Right.
And at Microsoft that I think of as a differentiator, which is how do you
innovate, but how do you do it at the speed of trust with governance?
I go back to that and though are people having copilot and other enterprise
generative offerings right now? It’s it’s a mix.
And I would say it’s actually a healthy thing right now because again, like
think about I think of A.I. tools for employees as a contact sport.
In fact, I think one of the like, you have to get in and use this stuff to
actually get things done. And that’s how I think of.
The challenge is amazing because and in fact, we did a study called the Work
Trend Index. And one of the things we found is this
class of companies, we call them frontier firms.
And the big distinction there is that they are not.
They’re not thinking of it as a snack. It’s a main meal, Right.
For their employees. It’s an organized adoption.
It’s not this thing off the side. They’re using multiple tools at
experimenting with things. But at the end of the end of the day,
they are also looking at what’s the order, why?
Where are they getting the governance and the security and as well as kind of
the latest access to intelligence. And that’s how I think about like where
Microsoft is playing. Wow.
Aparna, It’s been a great contact sport. I can’t play rugby anymore based on
contact sport. Yeah, different nature upon which I got
it. She’s Microsoft Chief Product Officer of
Experiences and Devices. Now we’ve got planning coming up telling
to the battle a contact sport, a chat pot dominance into big business that
we’re talking to the co-founders l.M Arena of Nexus Technology. Welcome back to a special edition of
Bloomberg Technology. We are live from the Bloomberg Tech
summit at the marina. How many jackpot arena?
Well, let us use this test run I models the project started by researchers at UC
Berkeley Sky Lab. It’s just raised $100 million in seed
funding Celsius. Angelopoulos Arena CEO and Wayne and Jan
Zilla, Marina, senior CTO, and they join us now.
And it’s great to have both of you gentlemen here.
Great to meet you. Thanks for having us.
For those who haven’t yet gone and played and battled out of the chat box
on your arena, what is it you’re offering?
So our marina is an open platform for evaluating and testing.
Well, that means that you can go to Ellen Marina Dot II and you can try all
the different, you know, top and top models from different providers.
That means Project Gemini, Claude, all the models that you might know the best
ones. And not only can you try them, but you
can also battle them against one another to see who wins.
And if you so choose, you can vote for the one that wins.
On your question, on your, you know, real use.
And then we use that data in order to construct leaderboards and evaluations
of these models. So you can go to our Maria Tadeo and
click on the leaderboard button and then you’ll see a ranking of all the
different models according to our user base.
And our user base is pretty diverse and largest.
Millions of people come to this Web site and vote for these different models.
And so we have all sorts of categories coding, math, instruction, following
multi turn hard prompts. And so we ranked with all of these
there’s a point to all of that, which is that the users themselves are
developers. They might even be on a product team,
right? And they’re trying to assess which
underlying technology works best for that goal.
If I’m going to log on and do this, just walk me through the steps in mind, my
motivation to do so. Yeah.
So I think the as we know, when people develop
application on top of A.I., they are just so many choices right now across
two very different providers. Right.
And then it’s getting, you know, increasingly challenging to compare with
different eyes, to evaluate them, to understand and monitor them, and then
solve. One question we got from developers I
was asking, which is for my for my use cases and not as a platform to provide
you, you know, a platform for you to test, come here.
And then with all these different evaluation as a foundation for you to
understand which effort is best for you, you then become a business with $100
million in backing. What’s the business model here?
How do you start to charge people for the use case?
Because you’re bringing a lot of value. Absolutely.
Well, I think you can see based on our conversation, that there’s a lot of
value to be brought in terms of bringing reliable A.I..
Right. So that’s one of the biggest problems in
A.I. writ large, is we have all these models
that are all deterministic. You know, they’re outputting text.
We don’t know how to evaluate them, we don’t really fully understand them.
And they’re being used in really subjective and interesting ways.
And I think businesses around the world have the question which is best for me,
How can I incorporate A.I. into my business reliably?
And that’s the market that we’re going to go after, guys.
Why do you need $100 million? What are you going to do with that?
I’ll be running the platforms Expensive, quite expensive way.
Then, like from a is it a compute issue? Like you are literally going to go out
and get yourself a nice cluster of Nvidia GPUs.
What are you going to do? Yeah, so currently we are serving
already millions of users a month and then,
you know, with all of these top ICE models from providers, they are not
cheap, right? And then we are focusing on like scaling
the platform as well. So I’ll bring it up to, you know, five x
ten x more usage and then in that case will be like, you know, spending tens of
millions of inference on serving those, Right?
Yeah, that’s why we’re great. Can I ask about potential gaming of the
system and Celsius? Because there has been some criticism
that and developers are coming on and then putting test beds out there,
getting them use that and putting the best ones forward.
How are you ensuring that people are getting the true experience of each chat
bot and ensuring that you’re getting the right results on the voting?
So I think you have to zoom out and think about how is the leader work
constructed. So in order for a model to do well on
the leaderboard, what needs to happen is new people are coming every day and
voting for us. So the only way that you can go up is if
people vote for you. There’s nothing gaming that our
community and you know that new people and you know their new face and Reddit
had a baby. Is that right?
Is that kind of what we’re talking about here?
I like that. Yeah.
Hugging face and right in open or something.
Okay. It’s also uses this.
I totally agree. Yeah.
So I think that’s that’s the sort of end of the story there because what happens
is that, you know, no matter what happens for a model to get number one,
that means our users rated it number one.
Guys, I would love to talk about talent. You know, part of the raise, I’m
assuming, is because you need to hire aggressively.
We are in an environment where engineers, particularly those from
overseas, you yourself like me and Kara, you know, came to this.
Contrary to what you see phrase. How difficult is that?
And what’s the priority for hiring police?
Yeah. So we’ve been like focusing on hiring is
our top priority right now. We are expanding.
And then in terms of like international, you know, talent, for sure we are.
That’s like a you know, we are going to be hiring a lot from them.
And then the, uh, the environment we are from the sci fi female, Right.
And super supportive from Berkeley and from our lab.
All have like probably close to half of them are from international students.
And you’re confident that they can remain so?
We hope so. Yeah.
Yeah. A lot of the best students in our lab or
international for sure. A lot of the best science funding,
Brooke, is about international students. So we’re hoping to keep them around the
team from our lab arena doing battle. There’s been a heavy battle and bruising
theme today when it comes to at least like the tool side of of generative AI
and and models and deploying them. Wow, what a fast start.
That does it for this very special edition of Bloomberg Technology Kara.
And we are live from Tech Summit in San Francisco.
We are you’ve got a great conversation coming up on stage on about the future
robotics. I go on all about the future of AI
investment. Stick with us.
This is Bloomberg Technology.