November. December was a movement from the hardware, from the infrastructure into the software and applications of AI.  Is that going to run on? That's right. Thanks, Caroline, for having me. I think in general, it should run on all

that said, we've come so far in such a short period of time this year with the AI software index being up 40% between early August and early December. So I think some consolidation of the gains we've seen is logical and I could

see more of that as we move through the beginning of the year for a few reasons. But I think in general the trend is up and I think we've reached an important point of demarcation for the space. Let's go into those few reasons why perhaps we're going to get a few

headwinds rather than tailwinds at the start. Is it macro policy? Is it Trump coming in? It's uncertainties over macro policy and Trump. And it's also just time-tested elements like we're entering the seasonally weaker part of the year.

Companies are likely to guide conservatively on their Q4 earnings calls, also with what we heard from the Fed this week. The group will likely have to adjust to higher interest rates for longer, and that has an impact on a few factors, including valuation.

And looking at valuation more broadly, there have been haves and have-nots. Some have been entering bid benchmarks. I think of Salesforce here, that has just been such a software winner. As people get galvanized by the generative AI opportunities, all those

valuations to heady, at what point do we start to see ultimately some of the P/E ratios look pretty healthy because they're managing to grow their earnings in lockstep?

Such a good question. Palantir is in somewhat rarefied air. It's the first name of scale to really

capitalize on generative AI in a large way. And it also has the government tailwinds of where its position. So it's doing tremendously well. Its valuation is head and shoulders above everyone else's. So I leave that alone for the rest of

the space. We are about to see, I think, much better growth than we've seen. The space has been digesting a flurry of buying activity during COVID, so it's been really two years of digestion and companies have been using technology as

a source of cost savings. But now with the advent of generative AI and especially agents, which I think market as a paradigm shift for the space, we're starting to see more investment and that's exciting. I mean, Palantir, just showing how much it's run up, it's currently trading

at 199 times future earnings. But let's go to where the oxygen is really being sucked out of the room: all Biogen, TKI. And again, I think there must be this question mark for investors and onlookers alike of, well, show me the money, show me the productivity, show me

how much this is actually going to affect my life, because I'm just hearing the words.

Right. Right. We're starting to see, which is so early. Salesforce is effectively leading this movement. So many people are they're working on

it, but they're one of the first ones out with true agile and tech functionality. I think it's been down seven weeks, maybe eight weeks. Have you used it? I personally have not used it. I've talked with a bunch of customers

who have been using it, who have been testing it in beta, and the feedback is excellent. The feedback is really compelling on accuracy, ease of use, and the ability to draw together critical enterprise data and make some use of it. So I think we're just on the cusp of

that. We're going to see other companies do that, but it makes for a compelling story, and we can start to dream of a better time where growth is better, and part of that growth is going to come not just from the events themselves, but from the underlying infrastructure.

We're seeing customers take on that hard work of improving their data, harmonizing it, getting it ready for use, and investments in cloud and underlying infrastructure to make that work. And we had the Databricks CEO raising a lot of money in the private markets in that role, about sort of being able to

use the data underlying. To that end, though, we're having a lot of startups talking about Gen2, trying to get in the space, taking on the stalwarts, the giants such as Salesforce. Are we going to see more M&A activity?

Once again, everyone's wanted Benioff to hold off, but more broadly, these software companies are going to get back into M&A with perhaps a more friendly administration. Sure. Well, yes, no, and yes, I think we absolutely do. I think part of the reason why we've

seen that rise in software valuations since the summer is because of this friendlier stance we're likely to see, at least on large companies buying smaller companies, which will leave the rest of large company administration aside. But there's that. We will see a lot of startups trying to

get into the space and some will be successful. There's always disruption when there's a new cycle, a new tech wave. But what is compelling is that we can see these large companies capitalize, augment their portfolio. We're seeing them do that.

Salesforce specifically, they are making tuck-in acquisitions. Ultimately, we probably see them do something, but right now they seem so focused on what they have at hand that we may not see what we've seen traditionally from the company in terms of large M&A.

And in fact, the company still has a large swath of...it's on the board who are highly engaged and keeping a close eye on the purse strings.

More broadly, therefore, when 2023, 2024 was so dominated by one name, Nvidia, and everyone asking you about the

infrastructure to the software names that you've really kept an eye on. Are we going to see this shift now? Is everyone going to just be more around agents? Is it going to be the time of Quantum? What do you think 2025 is going to be the cocktail party question for you?

Hmm. Okay. That's an exciting one and one that will change every week. We have cocktails for more. We've had this push and pull between semis and software the entire time as investors try to identify the real beneficiary. In reality, over time, they'll both be

strong beneficiaries in one way or another. These are the names within that group may shift. But I think software is starting its period of appreciation of AI, wheels in agents. As I mentioned, I think they represent the period of shift from homegrown AI or do-it-yourself AI to 'do it for me'.

AI, and the customers would prefer the vendors do it for them. It's got to be better, faster, cheaper in the end. And so I think that's a story that continues. And there are other things I like about

software, semis, or are in the process of getting beaten up. There's a lot of debate over whether we're going to see the big behemoth, million-GPU pre-trained models, or the smaller models, or post-test time training where the training occurs at the point of inference, and that will

have financial implications. But in the end, we're going to need a lot of semis for what we're about to do.

ClearBridge's Frisch: Salesforce Leading in Agentic AI

Welcome to the Bloomberg Technology YouTube channel. Subscribe for original series and mini-docs on game-changing technologies, profiles on innovators, live conversations with experts, and breaking technology news & analysis from Bloomberg TV.

ClearBridge's Frisch: Salesforce Leading in Agentic AI

ClearBridge's Frisch: Salesforce Leading in Agentic AI

Scripts