19:13
The midterm election results are in, and the AI industry could face a new wave of resistance to building data centers, or "compute campuses, " as OpenAI plans to rename them. Also this week, Squarespace and Canva are cutting their teams amid slowing growth, and we're tracking several big deals, including Marvel's $ 115 billion-and that's billions with a capital "M"-acquisition of Momentum. Stay with us. Hello, it's Equity again, TechCrunch's flagship podcast about the business of startups. I'm Anthony Ha, and I'm joined by TechCrunch's transportation editor, Kirsten Korosek, and Sean O'Kane, our senior special projects reporter. Kirsten, you're not where we
usually see you. What is happening? Yes, I'm in Boston. We just held our last event of the year-Founder Summit. A completely different type of event than Disrupt. Very focused on early-stage entrepreneurs, even those with just ideas. We had a really cool pitch presentation. I interviewed the co- founder of Toast. A great one-day event dedicated to advice for future founders, which was cool. And in a day I'm going to Web Summit. So I won't be on the record next week, sorry you'll miss me. But I have other, much bigger news: this is Sean O'Kane's last broadcast.
Boo. I'm glad at least someone was upset. We are all upset. We are all sad. No, I'll miss you guys. This, this is a real bummer. This is my last day at TechCrunch and my last post, and it was, it was so much fun-to have the opportunity to sit with you and think about the news in a different way than we usually do. Sometimes I feel like we could rethink everything here, and so I will miss it a lot. So, basically, you're saying you're going to miss your weekly tech news therapy session? Is that what you mean? Okay, okay.
That's right. It's a therapy session combined with a news review, in front of many people who remain faceless to me Yes. The best kind of therapy. Well, you're moving to the Wall Street Journal, so congratulations. We can't wait to compete with you and see what you write about. But we should discuss this week's news and begin our therapy session. There is nothing to discuss this week Yes. There was an event like Election Day, and wow, it shook up the tech industry, and I say that in a broad sense because it's really massive.
You know, Sean, I want to give you the floor because this is your last show, and also because you have a lot to say about what we're seeing in the first few days after Election Day and how that's impacting the AI industry and, in particular, data centers. Yes, I think this narrative began to take shape even before the elections, as well as in the days that followed them. There are a number of points, well, first, let's talk about the fact that one of the largest data center developers was also the largest donor to the Republican Party-Elon Musk. And, you know, it seems like he got almost nothing out of the half a billion dollar donation, maybe even more when we see the final numbers.
Moreover, now all of his businesses are in a state that has become as "blue" as it can be for the first time in a very long time. So I think that's the most important aspect in this discussion, but there have also been numerous initiatives across the country that could accelerate the construction of data centers. Almost all of them were rejected or did not receive support on the vote. And now the question arises: when new leadership comes in, democratic leadership that tends to "slow down " processes, and we can discuss the whole spectrum of what that will mean, right Some don't want it at all, and some simply require more caution and information before building anything. With them in power,
there is a consensus in the tech world that this will mean a lot of disruption for data centers, especially in places like Texas, where Governor-elect Inohosa has been quite open about this as a campaign promise she wants to focus on So yeah, this is something that I think the tech world is maybe even a little too focused on. I'm curious what you think about this, but it's definitely a major topic after Tuesday's events. I don't necessarily think they're overly focused on that, but I understand why it might seem a little disproportionate, because, you know, on data centers-there were definitely a number of initiatives, but it wasn't an issue that was put to a vote everywhere.
Um, and in different races the dynamics could be a little different. But there was a certain undercurrent and a sense that this was a nationwide issue, largely bipartisan, but the Democrats seemed to be able to harness that energy more effectively. And of course, there are probably many reasons why Democrats performed so well on Tuesday, but it seems that anti-AI, and especially anti-centrist sentiment, played a significant role. And that looks like a pretty powerful political force, and I suspect that the next two years, and maybe the period after that, will be more challenging for technology companies in general and for hyperscalers and data centers in particular.
I also suppose they must be a little concerned about whether this will remain an issue not just for those elected this year, but potentially in 2028 and beyond. Yes, definitely. I don't think the data center issue is what brought everyone to the polls and led to the so-called "blue wave." However, I will say that data centers are becoming an increasingly important grassroots issue, depending on the locale, in part because of how hyperscalers and companies approach the creation of these centers: very secretively, often requiring elected or
appointed officials to sign non-disclosure agreements; without sharing information about the scale of the project in environmentally sensitive areas, particularly in the Southwest, where there is a major problem with water scarcity. So I think a lot of people- Republicans, independents, and Democrats- think it's wrong, and that's what motivated them, but I don't think that was the only issue. But here's what's interesting. There's been a lot of talk about how we can operate in different circumstances, and we're already seeing companies like OpenAI suddenly say, "Oh, we can actually change our policies."
They just released what they call a "computational campus policy plan," and guess what-they will no longer require signing non-disclosure agreements. Um, they're going to share their plans and supposedly work less secretly. So I think what this shows is that when you create regulatory certainty in the market, or at least signals of movement in that direction, corporations respond. If you don't, they won't necessarily do what we hope. Right. I want to emphasize what you just said: part of what's happening, beyond defining these policy changes, is trying to rebrand data centers, because that term has become very toxic, so they're not data centers anymore
These are computing campuses, and I'm wondering if any of you think this will catch on. I will also note that one of the other very small effects of the election that I think will confirm my prediction on Equity a few weeks ago is this whole term " superintelligence." I think it's disappearing because I don't think any AI leader would be particularly inclined to follow Donald Trump's executive orders right now. So I feel like my prediction will soon be confirmed on this front. Looks like you just got a nod from one of Sean's puppies. My dog. Yeah, I mean, listen, computational campuses-these are the places where superintelligence is created, Anthony. Do n't you see how it's all connected? It just works that way. Well, yeah, I
mean, it's funny. Well, come on. This won't change anything. This will only make them more effort. For a company that likes to say it no longer focuses on side projects, OpenAI still really likes to move in different directions. And it's just another thing that makes you think: why are you even wasting your time on this? You know, this is unlikely to significantly change the situation. They have other more important things to do. I think it's a waste of time to try to make people think about it. Great idea for OpenAI: in the spirit of making threatening things cute, they should create some kind of cute avatar of a computing campus and represent it instead of an energy- and water-wasting data center.
I'm sure it will happen very soon. I mean, they could build more of their own offices, and if these are just great computing campuses, they should build their own office space there, or work next door, and see what happens. I think what I'm also interested in is this: how much the election is perhaps the second part of the double whammy that we seemed to be referring to last week, but which became a little clearer shortly after the recording. There's a significant shift in how Wall Street is reacting to many of these hyperscalers and AI companies, and a sense that they're not thrilled about the idea of spending so much money building new data centers and computing power.
So in a sense, there is now a financial incentive not to build all these data centers, and I'm very curious how that will affect the situation, because it will probably become more difficult to build such facilities due to regulation. Wall Street probably won't encourage you to build data centers, but at the same time, to achieve the desired scale, they will have to build them somewhere else. Yes, look up, you know? Look up at the night sky. We've talked a lot about the fact that, aside from the technical complexity of putting data centers in space, the attractive qualities of this, especially for someone like Elon Musk, are that there's
no "not in my backyard" traffic in space There are no environmental regulators who will come after you if something leaks or weathers out of your space data center. There are many reasons why companies like SpaceX are considering this idea, or Google, which launched its first, fairly small test, starting the Sunchaser project.
The thing is, with regard to my earlier comment that data centers may be taking up too much of the attention of the tech world and tech executives, I think that. Look, even if the ground-based data centers are shut down or slowed down to impossible deadlines and the focus shifts to orbital centers, so be it, but there's a long delay between now and then, and in the meantime, there are a lot of challenges waiting for big tech
I think we are in for a whole series of hearings where these leaders will be forced to answer to a Democratic-controlled Congress. They're going to have to account for, you know, the money they paid to the Trump administration and the inaugural committee, and they're going to have to account for the activities of Doge, which actually never got a proper detailed report on everything that the Doge representatives did, so there's still a lot of events to come before we see the results on the data centers, and the last thing I'll say is that
the market is confirming it, because CoreWeave and other companies have been down since Tuesday, their debt doesn't look the best, although Anthropic just got one of the largest debt packages. SpaceX just took on about $40 billion more in debt. So, there is definitely hedging going on in the market, especially with AI and data center focused companies, and I think these tech executives will have a lot to worry about in the near future. So, by the way, about the near future: yesterday, Meta Platforms CEO Dina Powell McCormick, at the Wall Street Journal technology conference, confirmed many of Anthony's thoughts about getting data centers becoming more difficult
She said something that I found very interesting: "I can say that every hyperscaler is now wondering, do we need to build outside the United States?" That was a pretty pointed comment, because if you think about it, all this push to build compute campuses, AI data centers, whatever you want to call them, revolves around the idea of the US winning the AI race and how important it is to have the infrastructure right here in the United States. So, going back to that passage about space, is that what she meant , or is she literally talking about placing them in Canada or other regions This reminds me a lot of another phase of my life, when I covered the energy industry, specifically oil and gas. And then I realized that there was a lot of
lobbying and complaints about the idea of regulating the industry. But once the industry became regulated, the only thing they really wanted was certainty in the market. And then they responded and mostly stuck to it. And I say "mostly" in quotes. Regulatory certainty actually matters in the market. And so I think that's what we were missing. It seemed like complete arbitrariness. And setting local standards and forcing companies to adhere to them, if they don't change every year or two, will allow them to adapt. And I don't think they need to go outside the United States or into space. This is my prediction. Um, that was my performance. We'll see how it all works out, and you can remind me of it next year if I'm wrong.
But there are other things going on in the industry, including, it seems, some personnel adjustments. I want to start by saying that Canva is laying off over 1,200 people amid slowing growth. I wonder if this is a consequence of what we've been seeing over the last year, or is something else going on? I've been thinking about this myself, because on some level it seems like part of a general trend where every tech company has been laying off a significant portion of its staff in recent years. This is usually somehow linked to artificial intelligence, there are debates about " AI-washing", or it is simply a correction of the consequences of over-hiring during the
pandemic. But it also appears that Canva and another company that recently made layoffs, Squarespace, both cut about 20% of their staff. They also seem related in a broader sense, as they are tools that have simplified the creation of images and graphics in one case, and websites in the other And of course, when you are the next step in this democratization or simplification, there is always the risk that someone from below will make it even easier and more accessible. So it makes sense to me that even if I don't think AI versions of, for example,
personal websites are necessarily better, they pose a serious threat to both companies. I know that in the case of Canva, cuts have occurred in all directions. Regarding your point about massive layoffs in the tech industry, I think we're still seeing echoes of what happened during and after the pandemic, when all these companies were hiring en masse. After emerging from the era of zero interest rates, people began to more soberly assess their real needs. You know, it's also interesting to think about the fact that for a while it seemed like a lot of other companies were, so to speak, following Musk's strategy on Twitter, when he
turned it into X and radically scaled everything down. It was tempting to think that he was setting an example that many other companies and executives would follow by resorting to such drastic cuts and laying off huge numbers of people. I think there were situations where that was indeed the case, but overall it's a different trend that we're seeing: maybe they just hired too much at the beginning, and a correction was inevitable. I, for one, was surprised that Canva even had that many people to cut, right? That's a lot of workers. Yes, I like your comment about this moment of "sobriety."
It doesn't have to be apocalyptic -just not to be a complete hype, a mindless waste of budgets, where everyone is constantly using tokens and just building, building, and building. And I truly believe that we are in for an adjustment. This is a pretty big change for these two companies, for sure. And I find it remarkable that they both launched their own AI-based products. Even here in Boston, at the Founders Summit, I talked to people- founders and business owners who are using AI tools-and they all said the same thing: they're excited, but now they're adjusting to the real world. There is
no longer this crazy pressure for everyone to use tokens all the time, instead the focus is shifting to smarter use of AI tools and what actually works. There was this very experimental, very rapid phase where we saw extreme growth. And I think a lot of companies right now, Canva and others, are taking stock and asking themselves: what can we really achieve and do we really need that many people to do it? I'm curious to see if other companies follow suit by the end of the year. This is the period, especially now or right after the holidays, when we see significant budget cuts and adjustments for 2027. For me, this is something we
will watch in the future to understand whether these are isolated incidents or a large-scale trend. One more point about Canva: and this is not to contradict Sean, because I think there is probably some broader hiring adjustment going on that these companies are making Canva also said that its revenue growth has slowed this year, and this has happened for the first time. You know, I don't know to what extent this is caused by artificial intelligence. And it 's worth noting that both of these companies also have their own AI-based products. But it's conceivable that in theory you could use a more
general-purpose chatbot or just an AI tool instead of something specifically designed for websites or art So I don't know exactly how much of this has to do with the emergence of such competing tools, but I wouldn't be surprised if it played a significant role. You can object to me. I'm about to leave. I think there is some of that, and I don't want to downplay the fact that more popular AI tools from OpenAI, Anthropic, and others are competing here in some way. I think this is clearly happening and that it will certainly put pressure on such companies in the future. However, there are two things I want to say about this.
First of all, I think there's a certain element here, and it's going to be very interesting. I'm interested to see how you'll cover this as the situation develops in this direction. The question arises: if Claude can do XYZ, then why do I need Squarespace? I think the answer to this question would be: Squarespace employs people. They offer certain things. You know, they create a finished product for you. And with Claude or, say, ChatGPT, you're constantly working in a mode of, "Yes, I can, I can create my own things for myself, " and that's great, they're becoming more and more functional, but there's no one person you can turn to with your problems, questions, or
advice, and while you can get some of that from a chatbot, not all of it. So these companies will just have to figure out how to create a product that covers all of these needs, and how to provide a service (which, in fact, they do) that will be there when you have questions or need something for what you're creating. And by the way, one of the things Canvas' CEO mentioned in her letter to all employees is that they will be trying to reduce " niche specialization," which means that employees can be asked to do more with less, presumably by relying on artificial intelligence tools like those being developed by Anthropic and OpenAI.
We see this in the corporate sector everywhere, and of course the question arises as to how far this can be taken; this moment is definitely present. I fully understand the value of using these tools to get more done than before, especially in programming. But now that we're talking about softer skills in the corporate world, there will be burnout from this; there are limits to what people using these tools can endure. So what balance will they be able to achieve, how much can they put on the shoulders of these workers before they break down - this is a very interesting question for the next few years.
This is an interesting point in terms of what this means for companies like Squarespace, and one can imagine a future in which some companies focus on " live" customer interactions to differentiate themselves from others. I'll be very interested to see how this develops, but I really want to discuss an incredible technology deal whose numbers just raised my eyebrows, and it's with a developer of specialized chips-a company called Marvel. They signed a deal to acquire Lumentum, all in stock, and it's, drum roll, $115 billion. That's a lot of money. What do you say?
Well, I had the same thought, that yes, this is really a lot of money. And I think my reaction to that probably reflects the transitional moment that we're in right now, where, I don't know, maybe a year ago this would have seemed normal because there were so many huge investments involved, big M&A deals. I mean, at any time this would have seemed like a very big deal, but things seem to have quieted down a bit now, especially for deals of this magnitude. Of course, there is some consolidation happening as people try to figure it all out and realize that maybe not all of these AI companies will be successful But seeing Marvel spend that much money, even if it's not in cash, is still a huge, serious commitment, and the number itself is impressive.
Yes, I want to explain the situation a little: this is one of the largest acquisitions in the history of technology, even if it is a stock-only deal, we count it as such, without any asterisks. Microsoft bought Activision Blizzard for $ 68.7 billion, and in 2022 Dell bought EMC for $67 billion. By the way, in 2016 it was a fairly large amount, it was quite a long time ago, but it gives an idea of the scale of such a deal. Yeah, I think one of the most interesting things about this deal is that it seems pretty indicative of what I was talking about all the time at Disrupt last month, particularly with Carl Almar from M13.
He lamented both on stage and behind the scenes how the inflated valuations of virtually any AI company-whether in software or, as here, hardware-absolutely distort the playing field when it comes to startup investments and acquisitions, and the tension between the two. And you know, his position as an investor was that they're missing out on a ton of opportunities to back companies that they think should be going through the early, mid, and late-stage funding stages, just because there's so much "air" stock that companies like Marvel can throw around, saying, "Hey, we see value in just acquiring this company, and we have the ability to do it right now on paper, so let's do it." So, there are all these distortions happening in this broader area
, and I feel like we're probably going to have to deal with that in the next few years. And I don't know what it's going to look like, but I don't think this is the last time we'll see something like this happen. Yes, this isn't the only acquisition we wanted to talk about this week. We're going to discuss other deals that I think would seem pretty big in a normal week, but they pale in comparison to the Marvel deal: CrowdStrike paid $400 million for cybersecurity company Fidler AI, and then $95 million for another company called Dreadnode. These deals seemed to be bundled into an announcement that CrowdStrike was making major investments to expand the scope of AI security, which seems very timely.
Yes, cybersecurity is so popular right now. I wonder why. Yeah, I mean, it's one of those things where I wonder what the process was like behind the scenes. Is this a case where talks were already underway, and then this, you know, " Black October" happened, and they accelerated, or what? I would like to know more about the pressure CrowdStrike was under to make such acquisitions. And because the cybersecurity situation in general is so uncertain, I think we're going to see a lot more of this kind of thing as consolidation happens around the
biggest names in the sector, because trust-it's so important when it comes to well, cybersecurity in general, that sounds like a silly statement. But the knowledge of companies like CrowdStrike, the knowledge of players, the presence of a reputation-all of that will have much greater value as we move into a world of greater uncertainty. This gives me an opportunity to correct something I said during our discussion of the PHP vulnerability a few weeks ago.
I already posted a fix in my post about this, but I want to mention it again because when people talk about that vulnerability, they say it was discovered by AI, which, as our colleague Zach Whittaker pointed out, is not true. It was discovered by human researchers participating in the OpenAI program. They used AI tools to investigate exactly how this vulnerability could be exploited, but this wasn't a case of AI just taking and finding this huge vulnerability on its own. It seems like that's the future we're heading towards, but I think all the factors you both talk about are still at play. There will be a lot of pressure on cybersecurity companies to have the technology and plans in place for how to handle similar
situations in the future. Indeed, leaders will have a great advantage in this market, because the trust and desire to have one supplier instead of ten different solutions look quite convincing. I completely agree. Thank you for making this correction. We forgive you. Thank you. It will be interesting to see what happens next. By the way, if anyone wants to know more details about what happened, we had Wiz's CTO on Equity last week, so check out that episode. Yes. You know, another big trend is microdramas. And that brings us to our final deal. I had no idea the numbers were so high. So this is one of those times where I thought, "Wow, wow." I need to reconsider my views on this. It turns out that over 118
million users watch microdramas in the series format every month. ReelShort (StoryTV) just raised $80 million in a Series C round from Hornbill Capital and NBC Universal. Who would have thought that you could make so much money on microdrama? I'm mostly just excited because, you know, one of the million roles I've had at TechCrunch is working as a media and entertainment reporter. And it feels like for so long, you know, every venture capitalist, and, you know, NBC Universal is a venture capitalist, but the idea of investing in a media startup was so ridiculous that, like, who would do that? There are so many much more profitable and promising areas for investment. Therefore, it is nice to see that
media companies can once again raise funds, albeit from other media companies, rather than from venture capitalists. Quibi is justified. Jeffrey Katzenberg was not wrong. He was simply ahead of his time. That's right. Well, you know what? This seems like the best time to end my latest episode of the Equity podcast. We, uh, it was nice doing this with you guys. Um, but you know, the show will go on and Equity Pod will be back next week. Stay tuned for Equity Pod at X and Threads, it was really cool. Thank you all.
Thank you. Goodbye, Sean, we will miss Equity is hosted by TechCrunch senior reporters, producer-Teresa Loconsolo, editor-Kel. Subscribe on YouTube or wherever you listen to podcasts, and stay tuned for upcoming events at techcrunch.com/events. Thank you very much for listening, see you next time.
