Bloomberg Audio Studios, podcasts, radio news. You're listening to the Bloomberg Intelligence podcast. Catch us live weekdays at ten am Eastern on Apple car Playing and broyd Otto with the Bloomberg Business App. Listen on demand wherever you get your podcasts, or watch us live on YouTube.
There's a story out there, and the headline doesn't help me because I really don't know what's going on here. Microsoft and Apple drop open AI board plans as scrutiny grows.
I thought we wanted the big.
Tech names to be supporting AI, and I don't know what's going on here.
So it was like a freebe scene. Isn't it like they're just watching?
Yeah, exactly, but they tell me it's important.
And Dina Bass, who's a really good technology she's been a great technology reporter for i'm gonna say thirty plus years. I was reading here stuff way before she came to Bloomberg. All right, So I'm gonna self admit I don't know what's going on here. So I said, please get me in deep Seeing, because I'm just guessing he knows what's
going on man deep Seeing. He's a tech analyst apt Bloomberg Intelligence, Microsoft and Apple drop plans to take board roles at open Ai in a surprise decision that underscores growing regulatory scrutiny a big text influence over artificial intelligence.
What's going on here?
Well, so, I mean if you roll back through last year when we had the open ai drama, you know, the board firing Sam Altman and then open Ai adding four new members, and I think everyone so far has been just sort of interested in how open ai has evolved since then in the sense that they have even though they have added independent board members. Microsoft, you know, and now Apple probably would have joined the board. It
doesn't make sense like when you have vested interests. In the case of Microsoft, they have a forty nine per sense steak in open ai and they're the primary cloud provider for open Ai. How can you do a good job of being an independent board member. And that's where you know, the regulatory scrutiny comes into play, because the regulators obviously are concerned about how Microsoft bought Inflection Ai. Even though it wasn't an acquisition, they got all the
employees of Inflection Ai. So that's when it comes down to the optics of it.
Here, break down for me again, what the relationship is now with open ai and Apple versus open ai and Microsoft.
So Microsoft has a forty nine percent stake in open Ai. They made a ten billion dollar investment, and so you could argue, if you're an investor in a company, you should get a board seat as a Sequoia and some other private investors that open Ai has. Apple, on the other hand, just partnered with open ai, which is going to be one of the providers of Genai model on their iOS and other devices, and so it's more of a customer supplier relationship if you want to call it
that way. That Apple is using open e as a supplier, why were they getting a board seat, Well, you could argue they could be an independent board member, like Fijisimo is one of the board members for open Eyes. She's
the CEO of Instacart. Now, I think that's where the fact that these technologies are so intertwined in terms of AI being deployed on Apple devices, you could argue the independence of the board comes into question, and given how important AI ethics and guardrails and all the things that
regulators care about that everyone cares about. Frankly, I think the independence of the board was questionable, and these companies decided to proactively not you join the board, which I think is good for optics.
So Microsoft is under scrutiny in Europe and by US regulators into I guess their alleged dominance of AI. What's the view on the street about that risk to Microsoft? And I don't know because at some point maybe Apple out of it, but certainly Microsoft.
I mean, the best way to look at it is these companies require a lot of compute for training their models. So whether it's open ai and Tropic, any of the independent GENI model providers, they need a lot of compute. Now, in the case of Microsoft, arrangement is really you know, full of their interests Because OpenAI ends up using Microsoft Compute for training their models.
What does that mean? What is compute?
So think of all the Nvidia GPUs that open ai needs, but it's deployed on Microsoft Cloud. So if I need compute to train the models, I'm not setting up my data center and Tropic open Ai they're not setting up data centers. They're using Microsoft Data centers to get access to that compute. And that's where they could have gone to Amazon or Google to get that compute. The reason they're going to Microsoft is because of the you know,
the stake they have and the board seed. And so if you are open ai and Microsoft is sitting in that board, are you really going to pick any of the alternative cloud providers for training your models? Probably not.
Do you feel like open ai will ever go public or be bought by Microsoft? Like, how does that situation play out?
I mean, everyone knows their antitrust concerns around Bigpeg buying any of these companies, So the acquisition is definitely out of the window. And that's why the way Microsoft got all the employees of Inflection Ai, it was a pseudo acquisition. You get all of the employees, you pay six hundred million dollars, and you're not calling it an acquisition. So in the case of OpenAI, obviously that's not gonna have happened. But they Microsoft does have a forty nine percent stake.
Google bought DeepMind for six hundred million. Now they bought it, you know, a few years back. But it just goes to show how, you know, these companies positioned themselves for the future and they end up buying a lot of the startups. So I don't see an acquisition on the cards. But my open AI is unique in the sense it's a nonprofit and then they are looking to build a business out of it. So their structure was sort of
very weird to begin with. And now they had that board trouble where it was a three percent board that fired their CEO. Now they are expanding with independent board members. So I don't think they've got it right still, but at least, you know, not having Microsoft and Apple on the board is good in terms of optics. And that's what I can conclude.
Here on surveillance.
Earlier today we had a London based headshunt which I had never heard from heard of rougher like I about thirty eight billion dollars in assets under under management.
They have a Parish call on the market.
They think so overheated, everything's overhead blah blah blah, and that includes AI. Now they're saying they believe AI that of course they've been wrong by their owned mission and they've been trying their peers because the market's been going up. But the AI call is simply that yeah, AI is transformed, blah blah blah blah.
But we don't even.
Know what the uses are and we don't all the money that's being invested in AI. Explain to me what the return on investment is on that. How do you answer that or deal with that?
I mean, we know some tangible use cases right now. One is from a developer co pilot perspective AI works, and this thing is getting deployed at every company. Developers are using copilot, same thing with customer service chatbots. This thing is productivity enhancing and this is getting deployed. So I think it's hard to argue that there is no tangible use case. But look, the valuations have run up and there will be a disconnect at some point where
valuations are way ahead of reality. And that's where you as an investor, you got to pick your spots. I mean, a TSMC will continue to benefit, whether it's in VideA, whether it's Apple or anyone else that needs the latest chips for AI, they won't continue to benefit.
And I think I can kind of write the press release from in Video or any company that's really been lifted by AI. You know, it's almost kind of a customer fatigue. They've already bought all this stuff, They've already invested all this money.
Our customers have pulled.
Three to four years of capex forward and spent it over the last twenty four months, and we're seeing instead of going forty eight percent, we're gonna go with thirty percent. That's a bad day, yeah, text and for the market, I think, And it's.
Still going to be uniform. So some sectors will see very tangible productivity benefit of AI, some others will have a delayed benefit or maybe no benefit, and that's where the disconnect will start to show up. But right now we are in that you know mode where everyone believes AI is the next big thing and it will have productivity boots.
And then just one quick point on what Paul was talking about too, just that the monetization part, Like you say, the use cases are there, but then how you monetize it? Yeah, Like we don't know what that really looks like yet, Like everyone's gonna keep wanting in video chips. Okay, I get that, But then what the use case is for that? Okay, then you got to monetize it. And at some point that's gonna have to be proven.
I mean, look at Cloud. So when Cloud came through the scene, everyone said, Okay, it's a secular trend. But the monetization came about over time, and there are certain cloud native apps that really did very well as a business. So that's where an AI native app I think it's going to do exceedingly well. Whereas if you're changing a business and you're overlaying AI, you may or may not be successful, but we'll find out.
Good time to be a tech analyst, he's the best.
Yeah, I know, like you can ask the dumbest questions and he's like right there for you. Yeah, I appreciate it.
It's amazing. Mandeep, thank you so much for joining us.
Man Deep seeing senior tech industryental from Bloomberg Intelligence coming into the New York office on like his management team, So.
We appreciate that. Thank you, man Man Deep.
You're listening to the Bloomberg Intelligence Podcast. Catch us live weekdays at ten am Eastern on applecar Play and Android Auto with the Bloomberg Business app. You can also listen live on Amazon Alexa from our flagship New York station Just Say Alexa playing Bloomberg eleven thirty ben.
Cher J Powell. He is speaking of the House Financial Services Committee. In his opening statement, we will bring you the Q and A when that kicks off. Steve Matthews' Bloomberg Federal Reserve reporter. He joins us, Now he's anything we're going to hear different today than yesterday.
I think the plan is not to hear anything different. I mean, Powell's plan yesterday was pretty clearly to try not to make news. And he was asked repeatedly about we're going to get a right cut in September and or July or whenever, and he said, I'm not going to give any guidance on rights. And I think he
was trying to be very balanced. In most of the recent testimony, the semi annual Hemphrey Hawkins testimony, he has tried not to make a lot of news because, you know, and this is a particular political situation, it's right before the election. He doesn't want to make a lot of news. So I think that you're going to see the same tone today, Alex.
One of the things I love about Bloomberg and Bloomberg News.
Is we actually have a dedicated reporter for the Federal Reserve.
How cool is that? Nobody else does that?
No, it's really cool.
Yeah, Steve, Yeah, you're the best. Steve, Thanks so much.
I want to ask you just mentioned the election help us understand how the FED thinks about timing of any rate movement and the election when when is too close?
Are you stealing John Tucker's question? He's totally stealing John Tucker's question. Sorry, go ahead, Steve.
There is a theory out there among Fed watchers that the FED would prefer not to move in September. It's the last meeting before the election. If they move in September, there will be complaints, Oh, you're doing it for political reasons. What Powell has said about this is they don't think about it at all. If September is the right time, they will move in September, and you know it doesn't enter their discussions or as an influence at all. So you know, some people believe there is a little bit
higher bar to move in September. But you know, if that's what the data says is the right time, that they would go ahead and move.
I'm going to bring in what Joe Wisenthal was talking about in his column today, saying that yesterday the headline was Powell says that labor is not a source of inflationary pressures now and his point was like, well, that's kind of the game, right, Like if the FED no longer sees the risks of that wage price spiral and that inflation spiral from the labor market, which is what the FED can control. If the FED looks at it that way, that could set him up for a cut. What do you think about that?
That's certainly true. I agree with Joe. Joe's a smart guy. Over the last three years, Powell has basically said, it's all about inflation. We have one sided risks. We want to return to price stability. The labor market is fine. Now he's saying there are two sided wrests. There's risks on inflation, there's risks on the labor market. And that does set up a rate cut. I mean, because you know they don't want to see the labor market softened further.
Steve to make a rate change, either up or down. Does it have to be unanimous or what does a FED like to do?
Typically it doesn't have to be unanimous. Powell has kept things unanimous for the last couple of years. But I think you know, particularly in the political environment, he would much prefer that there be unanimous vote. But there are a couple of hawks on the committee, so that that may be tough. To get. I mean, you probably will get a dissent.
What do you think we're going to get tomorrow with inflation? Like, what do you think the Fed? I mean JPL yesterday definitely talked about how it's still going to be inflation X services that core number, right, what do you think we're going to learn?
I mean, all the estimates are that the CPI is going to come in relatively low, and the Fed, you have to remember, is focused on PCEE inflation. CPI goes into it PPI, which comes out a little bit later. We'll also go into that, and after PPI sometimes CPI can be misleading. After PPI you'll get some estimates of what they're their preferred measure of inflation is for the month.
And but the estimates are things are coming down. Housing prices are coming down, and they know rents have been coming down, and that's delayed in the official data.
Steve, the Fed had a you know, a knock certainly that they were too late to raise rates here and that they really missed the boat there. Does that influence kind of they're thinking today or they just kind of sloth that off.
It does influence their thinking in the respect that you know they don't want to see a recurrence of inflation, and if they're going to air on one side or the other, they want to air on the side of not seeing a recurrence. So they're afraid that if inflation were to bark up again, that would be really problematic for inflation expectations.
All right, Steve, thanks so much. We really appreciate Steve. Matthews Feder reserve reporter for Bloomberg News.
This is the Bloomberg Intelligence podcast, available on Apple, Spotify, and anywhere else you will get your podcasts. Listen live each weekday ten am to noon Eastern on Bloomberg dot Com, the iHeart Radio app tune In, and the Bloomberg and Ess that. You can also watch us live every weekday on YouTube and always on the Bloomberg terminal
M