WEBVTT - Instant Reaction: Anthropic Files for IPO

0:00:02.560 --> 0:00:04.240
<v Speaker 1>Bloomberg Audio Studios.

0:00:04.700 --> 0:00:05.420
<v Speaker 2>Podcasts.

0:00:05.720 --> 0:00:06.240
<v Speaker 1>Radio.

0:00:06.680 --> 0:00:07.110
<v Speaker 3>News.

0:00:09.210 --> 0:00:12.090
<v Speaker 1>This is a breaking news update from Bloomberg.

0:00:12.950 --> 0:00:17.369
<v Speaker 2>Instant reaction and analysis from our 3,000 journalists and analysts

0:00:17.570 --> 0:00:18.470
<v Speaker 2>around the world.

0:00:19.530 --> 0:00:21.470
<v Speaker 1>A couple of red headlines crossing the Bloomberg terminal. I

0:00:21.480 --> 0:00:24.140
<v Speaker 1>want to get them to you. Anthropic has filed publicly

0:00:24.200 --> 0:00:26.180
<v Speaker 1>for its U.S. IPO. One of the things that's just

0:00:26.220 --> 0:00:28.920
<v Speaker 1>amazing to me on this Anthropic stuff, first half revenue, $ 16.

0:00:28.760 --> 0:00:29.160
<v Speaker 4>3 billion.

0:00:30.580 --> 0:00:34.120
<v Speaker 1>Versus $ 1. 2 billion last year. So big growth in revenue.

0:00:34.240 --> 0:00:39.470
<v Speaker 1>But first half net loss, $ 80. 7 billion. So wow, the

0:00:39.510 --> 0:00:41.529
<v Speaker 1>numbers are just amazing. Losing a lot more than they're

0:00:41.550 --> 0:00:44.370
<v Speaker 1>taking in, I think is the takeaway there. Exactly right.

0:00:44.510 --> 0:00:45.760
<v Speaker 2>The ticker symbol A-N-T.

0:00:45.840 --> 0:00:49.280
<v Speaker 1>A-N-T. Goldman Sachs, Morgan Stanley running this deal, as you

0:00:49.320 --> 0:00:51.640
<v Speaker 1>would expect. Let's get some more reporting on this news.

0:00:51.700 --> 0:00:55.160
<v Speaker 1>Peter Elstrom, Executive Editor for Global Technology for Bloomberg News

0:00:55.320 --> 0:00:59.650
<v Speaker 1>joins us here. Peter, the numbers are just staggering. Where

0:00:59.690 --> 0:01:03.510
<v Speaker 1>to start? What do you make about this highly anticipated IPO?

0:01:03.950 --> 0:01:07.060
<v Speaker 4>The S-1 filing just dropped with all the numbers, so

0:01:07.140 --> 0:01:09.319
<v Speaker 4>we're all pouring through it right now. I think you

0:01:09.360 --> 0:01:11.780
<v Speaker 4>flagged the numbers that people are going to start with.

0:01:11.860 --> 0:01:15.069
<v Speaker 4>It's the growth in the losses that they've had as

0:01:15.090 --> 0:01:17.810
<v Speaker 4>you mentioned the revenue for the first half was sixteen

0:01:17.830 --> 0:01:20.569
<v Speaker 4>point three billion dollars that was up from. One point

0:01:20.630 --> 0:01:24.130
<v Speaker 4>two billion- a year earlier but the losses are quite

0:01:24.190 --> 0:01:27.490
<v Speaker 4>staggering as you mentioned just over seventy billion dollars. That

0:01:27.530 --> 0:01:31.789
<v Speaker 4>includes some special charges- certainly it's not just operating losses

0:01:31.850 --> 0:01:33.990
<v Speaker 4>there but- as we look to the S one certainly

0:01:34.030 --> 0:01:36.229
<v Speaker 4>people are going to be digging into those details and

0:01:36.250 --> 0:01:40.429
<v Speaker 4>see exactly- how they've managed those- that financial performance.

0:01:41.040 --> 0:01:44.140
<v Speaker 1>Peter, I'm looking at SpaceX, which is going to be

0:01:44.180 --> 0:01:47.030
<v Speaker 1>obviously a comp for this IPO, if for no other

0:01:47.060 --> 0:01:50.910
<v Speaker 1>reason than the size here. SpaceX IPO at 135, stocks

0:01:50.970 --> 0:01:53.870
<v Speaker 1>now trading at 147. That's a good thing if you're

0:01:53.890 --> 0:01:56.450
<v Speaker 1>trying to bring another mega deal. What do you think

0:01:56.490 --> 0:02:00.150
<v Speaker 1>the strategies of this management team is when they get

0:02:00.210 --> 0:02:02.450
<v Speaker 1>on the road here? Because it's such a big deal here.

0:02:03.070 --> 0:02:05.490
<v Speaker 4>It's an enormous deal, and certainly the markets have been

0:02:05.530 --> 0:02:10.510
<v Speaker 4>a bit skittish about AI investments of late. Certainly, we've

0:02:10.550 --> 0:02:13.390
<v Speaker 4>seen that with CoreWeave in particular, that on the credit side,

0:02:13.430 --> 0:02:16.350
<v Speaker 4>there are some growing concerns about how these companies are

0:02:16.370 --> 0:02:18.850
<v Speaker 4>going to be able to execute of course anthropic sits

0:02:18.889 --> 0:02:22.260
<v Speaker 4>in a different uh... part of the value chain uh...

0:02:22.230 --> 0:02:25.709
<v Speaker 4>with their frontier model um... clawed getting lots of revenue

0:02:25.750 --> 0:02:29.030
<v Speaker 4>generating lots of revenue uh... at this point uh... they

0:02:29.070 --> 0:02:31.470
<v Speaker 4>have a message at the top of the s one

0:02:31.510 --> 0:02:33.870
<v Speaker 4>where dario on the day the c e o his

0:02:33.910 --> 0:02:37.490
<v Speaker 4>sister daniella daniella uh... in the rest of the founding

0:02:37.590 --> 0:02:40.060
<v Speaker 4>team lay out their principles for this and certainly they're

0:02:40.070 --> 0:02:43.500
<v Speaker 4>talking about a eyes one of the most impactful technologies

0:02:44.380 --> 0:02:46.620
<v Speaker 4>of our generation, I think they're going to try to

0:02:46.660 --> 0:02:49.820
<v Speaker 4>get investors excited about the opportunities that lie ahead of us,

0:02:50.160 --> 0:02:52.920
<v Speaker 4>that this is not just about building out this infrastructure,

0:02:52.960 --> 0:02:55.540
<v Speaker 4>but then using that technology to solve a bunch of

0:02:55.600 --> 0:02:57.400
<v Speaker 4>other problems that we see in the future. And they

0:02:57.700 --> 0:03:02.320
<v Speaker 4>articulate with a very substantial opening memo in the S-1.

0:03:02.380 --> 0:03:04.500
<v Speaker 4>There are risk factors, though. We probably want to talk

0:03:04.520 --> 0:03:05.260
<v Speaker 4>about those, too.

0:03:05.280 --> 0:03:06.760
<v Speaker 2>Yeah, no, tell us about that, then.

0:03:06.800 --> 0:03:07.740
<v Speaker 1>What are the risk factors?

0:03:08.700 --> 0:03:10.660
<v Speaker 4>Certainly, there are pages and pages of risk factors. It

0:03:10.720 --> 0:03:13.860
<v Speaker 4>looks like roughly 100, almost 100 pages of risk factors.

0:03:13.880 --> 0:03:15.649
<v Speaker 2>They've talked about this in the past.

0:03:16.190 --> 0:03:19.170
<v Speaker 4>Yeah, certainly. We talked about this a couple months ago

0:03:19.210 --> 0:03:23.299
<v Speaker 4>when they had one of their researchers resign and warn

0:03:23.600 --> 0:03:26.740
<v Speaker 4>about how the Frontier Labs were gambling with human lives.

0:03:27.290 --> 0:03:31.390
<v Speaker 4>So certainly they are raising questions about the ability to

0:03:31.450 --> 0:03:34.530
<v Speaker 4>control AI, what it's going to do for the economy,

0:03:34.570 --> 0:03:36.690
<v Speaker 4>for job losses, for all these things that we've heard

0:03:36.730 --> 0:03:39.350
<v Speaker 4>over the past few weeks. And they articulate that in

0:03:39.450 --> 0:03:42.250
<v Speaker 4>quite a bit of detail in the S-1.

0:03:42.750 --> 0:03:45.130
<v Speaker 1>Peter, what do we know about the timing here, when

0:03:45.170 --> 0:03:47.550
<v Speaker 1>they might start a roadshow, when they would like to

0:03:47.590 --> 0:03:49.350
<v Speaker 1>maybe price this thing? What do we know so far?

0:03:50.160 --> 0:03:51.920
<v Speaker 4>Well, we've heard so far that they do want to

0:03:51.940 --> 0:03:56.100
<v Speaker 4>get this done before Thanksgiving, so the window is closing there.

0:03:56.700 --> 0:03:59.620
<v Speaker 4>There's a period of time where investors get to digest

0:03:59.660 --> 0:04:03.210
<v Speaker 4>this information. There'll be a roadshow where management will go

0:04:03.320 --> 0:04:06.260
<v Speaker 4>out and talk with investors. Certainly, the investment bankers are

0:04:06.280 --> 0:04:08.970
<v Speaker 4>going to play a central role in how those conversations

0:04:09.050 --> 0:04:11.850
<v Speaker 4>go and how they articulate the message there. They'll meet

0:04:11.870 --> 0:04:14.290
<v Speaker 4>with investors, answer their questions, and then the goal will

0:04:14.310 --> 0:04:16.110
<v Speaker 4>be to float after that. All right.

0:04:16.430 --> 0:04:18.529
<v Speaker 1>I've written a lot of prospectuses in my day, and

0:04:18.589 --> 0:04:22.630
<v Speaker 1>I never had a risk section that includes this. Anthropic

0:04:23.050 --> 0:04:27.750
<v Speaker 1>warns of catastrophic existential risk to humanity. How do you

0:04:27.770 --> 0:04:28.830
<v Speaker 1>price that one?

0:04:29.510 --> 0:04:30.409
<v Speaker 2>I don't know, Jess.

0:04:30.730 --> 0:04:32.650
<v Speaker 1>How does that work? I like the way Bloomberg News

0:04:32.670 --> 0:04:36.810
<v Speaker 1>characterizes it here. Anthropic files for IPO to become first

0:04:37.029 --> 0:04:41.289
<v Speaker 1>frontier AI lab to list. Going first, that's big. Mandeep

0:04:41.330 --> 0:04:44.150
<v Speaker 1>Singh joins us here. He's head of all the technology

0:04:44.190 --> 0:04:48.270
<v Speaker 1>research for Bloomberg Intelligence. Mandeep, what's the story you think

0:04:48.830 --> 0:04:51.550
<v Speaker 1>Anthropic needs to tell when they start the roadshow? What

0:04:51.570 --> 0:04:52.740
<v Speaker 1>does the street really need to hear?

0:04:53.220 --> 0:04:55.760
<v Speaker 3>I think they have a really nice chart in the

0:04:55.800 --> 0:05:00.529
<v Speaker 3>S1 on compute efficiency, which is efficiency, which is basically

0:05:00.589 --> 0:05:05.550
<v Speaker 3>saying for every dollar of compute. we have a much

0:05:05.610 --> 0:05:08.510
<v Speaker 3>better trend in terms of revenue that we are able

0:05:08.550 --> 0:05:12.250
<v Speaker 3>to generate. In other words, that's a proxy for gross margins,

0:05:12.730 --> 0:05:17.170
<v Speaker 3>and gross margins are inching closer to 50%, whereas just

0:05:17.230 --> 0:05:21.620
<v Speaker 3>a quarter back, it was like sub 20%. So clearly,

0:05:22.100 --> 0:05:26.380
<v Speaker 3>they are showing that even though this is a variable cost,

0:05:26.560 --> 0:05:30.220
<v Speaker 3>high variable cost business, we are able to improve our

0:05:30.260 --> 0:05:33.630
<v Speaker 3>margins and that's driven by the scale that they have.

0:05:34.110 --> 0:05:37.270
<v Speaker 5>Well, Paul and I were talking about how within the filing,

0:05:37.510 --> 0:05:41.710
<v Speaker 5>it was talking about risks, how the catastrophic existential risks

0:05:41.850 --> 0:05:44.690
<v Speaker 5>to humanity. How do you price something like that when

0:05:44.770 --> 0:05:46.360
<v Speaker 5>it's in this type of report?

0:05:47.010 --> 0:05:48.960
<v Speaker 3>Well, I mean, what they are saying, and this is

0:05:49.000 --> 0:05:51.720
<v Speaker 3>very typical of, you know, a risk disclosure for any

0:05:51.820 --> 0:05:55.520
<v Speaker 3>new technology IPO. They will say, you know, a worst

0:05:55.560 --> 0:05:58.820
<v Speaker 3>case scenario is where things, you know, get out of

0:05:58.870 --> 0:06:03.029
<v Speaker 3>hand just to make sure that, you know, they avoid liabilities.

0:06:03.570 --> 0:06:06.410
<v Speaker 3>in terms of something like this, where you are developing

0:06:06.450 --> 0:06:11.250
<v Speaker 3>a model. We know these models have had cybersecurity issues.

0:06:11.510 --> 0:06:15.789
<v Speaker 3>And I think it's very typical of any risk disclosure

0:06:16.010 --> 0:06:18.270
<v Speaker 3>of a new technology company going public.

0:06:18.970 --> 0:06:21.390
<v Speaker 1>Again, I was just looking at the Bloomberg News reporting

0:06:21.410 --> 0:06:25.110
<v Speaker 1>that maybe they've talked down their valuation of this company

0:06:25.270 --> 0:06:27.859
<v Speaker 1>a little bit. Does that just reflect Kind of the

0:06:28.440 --> 0:06:31.219
<v Speaker 1>tech market, maybe the AI story getting a little bit

0:06:31.260 --> 0:06:33.560
<v Speaker 1>of pushback over the last several weeks, months?

0:06:33.960 --> 0:06:38.670
<v Speaker 3>Yes, I think the whole backdrop isn't as good as

0:06:38.730 --> 0:06:42.230
<v Speaker 3>it was maybe three, four months back. And SpaceX probably

0:06:42.330 --> 0:06:46.550
<v Speaker 3>timed it better in that perspective. But to my mind,

0:06:46.630 --> 0:06:49.979
<v Speaker 3>I think a company like Anthropic, when it has such power,

0:06:50.310 --> 0:06:54.740
<v Speaker 3>fast growth with improving margins, it's a good story.

0:06:54.279 --> 0:06:55.779
<v Speaker 2>To go out to investors with.

0:06:55.839 --> 0:07:00.060
<v Speaker 3>So even though the bankers may price the IPO conservatively,

0:07:00.080 --> 0:07:03.060
<v Speaker 3>I wouldn't be surprised, you know, that's to make sure

0:07:03.120 --> 0:07:04.659
<v Speaker 3>that the IPO is oversubscribed.

0:07:05.300 --> 0:07:07.250
<v Speaker 5>What do you think this tells us to more broadly

0:07:07.310 --> 0:07:09.640
<v Speaker 5>about some of these other delayed IPOs that have been

0:07:09.670 --> 0:07:11.930
<v Speaker 5>in the pipeline as far as just like the demand

0:07:11.970 --> 0:07:12.830
<v Speaker 5>for something like this?

0:07:13.170 --> 0:07:16.250
<v Speaker 3>Yeah, you need the scale of someone like Entropic to

0:07:16.290 --> 0:07:18.870
<v Speaker 3>go out to investors with right now. If you are

0:07:18.930 --> 0:07:22.950
<v Speaker 3>a subscale player where you don't have an improving margin

0:07:23.010 --> 0:07:26.170
<v Speaker 3>story along with top line growth, That's when you pull

0:07:26.210 --> 0:07:28.910
<v Speaker 3>back your IPO. And we saw that with a few

0:07:28.970 --> 0:07:31.590
<v Speaker 3>of the smaller names where I bet you the margins

0:07:31.970 --> 0:07:35.530
<v Speaker 3>weren't really improving, whereas Entropic can do it because they

0:07:35.550 --> 0:07:38.750
<v Speaker 3>have the scale. And they are obviously, you know, the

0:07:38.790 --> 0:07:41.420
<v Speaker 3>poster child when it comes to this new business model

0:07:41.950 --> 0:07:42.770
<v Speaker 3>of LLMs.

0:07:43.390 --> 0:07:47.360
<v Speaker 1>For most IPOs, one of the critical factors is management

0:07:47.780 --> 0:07:49.870
<v Speaker 1>because you don't know them. And you kind of got

0:07:49.910 --> 0:07:51.730
<v Speaker 1>to make an assessment of the quality of management. What

0:07:51.750 --> 0:07:53.310
<v Speaker 1>do we know about this management team?

0:07:54.310 --> 0:07:55.190
<v Speaker 2>Not a whole lot.

0:07:55.270 --> 0:07:57.210
<v Speaker 3>And I think you bring up a very good point

0:07:57.250 --> 0:08:02.220
<v Speaker 3>that they don't have the track record as someone in

0:08:02.260 --> 0:08:04.780
<v Speaker 3>terms of having seasoned executives and whatnot.

0:08:04.940 --> 0:08:06.840
<v Speaker 2>Although they did bring in a new.

0:08:06.760 --> 0:08:10.920
<v Speaker 3>CFO a few months back who I think was front

0:08:10.960 --> 0:08:14.540
<v Speaker 3>and center with this IPO process. And I do think

0:08:14.740 --> 0:08:18.650
<v Speaker 3>it made a difference in terms of the disclosures around

0:08:18.690 --> 0:08:23.330
<v Speaker 3>that $ 750 billion in commitment number. How does that translate

0:08:23.390 --> 0:08:27.140
<v Speaker 3>into the income statement and what sort of obligations they have?

0:08:27.540 --> 0:08:29.960
<v Speaker 3>So from that perspective, I think they do have a

0:08:30.020 --> 0:08:33.240
<v Speaker 3>steady hand in terms of the new CFO that they

0:08:33.280 --> 0:08:33.800
<v Speaker 3>have brought in.

0:08:34.080 --> 0:08:36.300
<v Speaker 5>So what's the next thing you're watching for after this

0:08:36.360 --> 0:08:37.230
<v Speaker 5>report just dropped?

0:08:38.010 --> 0:08:39.810
<v Speaker 2>It has to be the IPO pricing.

0:08:39.970 --> 0:08:44.429
<v Speaker 3>And the bankers will price the IPO within, I would say,

0:08:44.429 --> 0:08:45.620
<v Speaker 3>10 days or so.

0:08:46.240 --> 0:08:48.080
<v Speaker 2>And then the roadshow will begin.

0:08:48.160 --> 0:08:51.580
<v Speaker 3>So to my mind, where they price the IPO, to

0:08:51.620 --> 0:08:55.100
<v Speaker 3>Paul's point, if they're too aggressive, then there is a risk,

0:08:55.140 --> 0:08:56.880
<v Speaker 3>you know, there could be disappointment.

0:08:56.960 --> 0:08:58.480
<v Speaker 2>Although I think going.

0:08:58.300 --> 0:09:02.250
<v Speaker 3>By the rumors, they're going to price it conservatively so

0:09:02.290 --> 0:09:05.030
<v Speaker 3>that you have a big first day IPO pop.

0:09:05.150 --> 0:09:06.230
<v Speaker 2>And I think that will be good.

0:09:06.490 --> 0:09:08.380
<v Speaker 1>All right, Mandeep, thank you so much. We appreciate it.

0:09:08.390 --> 0:09:12.380
<v Speaker 1>Mandeep Singh, nobody better to chat up this anthropic S1

0:09:12.640 --> 0:09:14.710
<v Speaker 1>than Mandeep Singh. He runs all of our tech research

0:09:14.910 --> 0:09:15.890
<v Speaker 1>at Bloomberg Intelligence.