The Wall Street Journal
WSJ:
WSJ's Take On the Week
SUNDAY, FEBRUARY 8, 2026
2/8/2026 6:00:00 AMShare This Episode
Why This VC Says AI and Robotics Will Put Every Human Job at Risk
「なぜこのベンチャーキャピタリストは、AIとロボットがあらゆる人間の仕事を脅かすと言うのか」
In this week's episode of WSJ’s Take On the Week, our hosts Telis Demos and Miriam Gottfried get into why the market still hasn’t made up its mind on Kevin Warsh’s nomination as the Federal Reserve chair, and why gold and silver trades fell on the news. Then Telis says he’ll be looking at Coinbase and Robinhood’s earnings this week to help make sense of bitcoin's falling value. Our hosts then break down the software selloff that followed Anthropic’s release of new legal tools.
After the break, Miriam is joined by Hemant Taneja, CEO of General Catalyst, at WSJ Invest Live. They discuss his investment philosophy in the age of AI. Taneja explains why he waited for a $60 billion valuation to invest in Anthropic and shares his view on why market bubbles can actually be a force for good. He also provides his view on how AI and robotics could challenge every human skill within the next 20 to 30 years.
This is WSJ’s Take On the Week where co-hosts Telis Demos, Heard on the Street’s banking and money columnist, and Miriam Gottfried, WSJ’s private equity reporter, cut through the noise and dive into markets, the economy and finance—the big trades, key players and business news ahead.
- VC (Venture Capitalist) /ˈvɛn.tʃɚ ˈkæp.ɪ.təl.ɪst/ ベンチャーキャピタリスト（起業投資家）
- put every human job at risk /pʊt ˈɛvri ˈhjuːmən ʤɑːb æt rɪsk/ 人間のあらゆる職業を危険にさらす
- make up its mind /meɪk ʌp ɪts maɪnd/ （意思・判断を）決める
- trades fell on the news /treɪdz fɛl ɑːn ðə nuːz/ ニュースを受けて取引が下落した
- break down /breɪk daʊn/ 分解して説明する、分析する
- investment philosophy /ɪnˈvɛstmənt fəˈlɑːsəfi/ 投資哲学、投資の考え方
- market bubbles /ˈmɑːrkɪt ˈbʌbəlz/ 市場バブル
- cut through the noise /kʌt θruː ðə nɔɪz/ 雑音を切り分けて核心に迫る
- key players /kiː ˈpleɪərz/ 主要人物・主要企業
FEBRUARY 8a 2026
Mirriam Gottfried: Hi, Telis.
Telis Demos: Hi, Miriam.
Mirriam Gottfried: Good to see you again.
Telis Demos: Good to see you too, back here at WSJ's Take On The Week.
Mirriam Gottfried: I know. We had the privilege of spending a good amount of time together this past week at WSJ's inaugural WSJ Invest Live conference.
Telis Demos: It was in West Palm Beach. We were hoping for sunny Florida as relief from this snowmageddon gray of New York.
Mirriam Gottfried: But it was kind of cold.
Telis Demos: It was like, 50s, which I know to Floridians is like-
Mirriam Gottfried: It was in the 40s when I landed. That's pretty cool.
Telis Demos: I got off the plane and I felt like, oh, it's Spring-like around here.
- have the privilege of ~ing /ˈprɪvəlɪdʒ/ ～する光栄に恵まれる
- a good amount of time /ə ɡʊd əˈmaʊnt/ かなりの時間
- inaugural /ɪˈnɔːɡjərəl/ 第1回目の、初開催の
- conference /ˈkɑːnfərəns/ 会議、カンファレンス
- snowmageddon /ˌsnoʊməˈɡedn/ 大雪災害（snow＋Armageddon）
- gray of New York /ɡreɪ/ ニューヨークのどんよりした天気
- ※ 50s / 40s はすべて 華氏（°F） → 50°F ≈ 10℃、40°F ≈ 4～9℃
Mirriam Gottfried: Anyway, the conversations were very interesting so it was all worth it. And we spent a good amount of time inside anyway at the conference. But coming up in a bit, we'll get to hear my interview that I did with General Catalyst CEO, Hemant Taneja, all about AI and his investment philosophy around AI.
Telis Demos: Very relevant to what's going on in software. We'll get to that in a bit.
Mirriam Gottfried: Exactly. But first we want to talk about another big topic that came up a lot at the conference, which is our new Federal Reserve chair nominee, Kevin Warsh.
Telis Demos: So now the debate is, okay, well, what's Kevin Warsh going to do?
Mirriam Gottfried: What does that mean? Yeah.
Telis Demos: Obviously Donald Trump really would like rates to be super low. The question about Fed independence, is he going to name somebody who's definitely going to do that? And so we asked people, we actually asked, so the audience at WSJ Invest Live was lots of money managers, people in the markets, people in the business, wealth advisors. And Gunjan Banerji, who I co-hosted a panel with about this, put a little straw poll to the audience. And she said, "Show of hands, who thinks Kevin Warsh is a hawk that's going to try and raise interest rates to control inflation? Who thinks he's a dove?" And so you would think that the answer would be, "Well, obviously he must be a dove if Trump-
- straw poll /strɔː poʊl/ 簡易・非公式アンケート
- show of hands /ʃoʊ əv hændz/ 挙手
- hawk /hɔːk/ タカ派（金融引き締め重視）
- dove /dʌv/ ハト派（金融緩和重視）
Mirriam Gottfried: Because that's what Trump wanted.
Telis Demos: ... named him, right"? But here's the thing. First of all, not a lot of people put their hand up for either answer, which is I think interesting.
Mirriam Gottfried: People don't know. Yeah.
Telis Demos: And the few hands that went up, a couple more went up for hawk, which really gets into the history of Kevin Warsh, which is that there have been times when he has been very concerned about the expansion of the Fed and interest rates. Specifically, he thinks that the expanded balance sheet has pushed up inflation. He's talked about lowering that balance sheet. Now, there have been times when he's also talked about maybe rates should be higher. Now he says that maybe rates should be lower. If we shrink the balance sheet, that takes pressure off inflation, then rates can be lower. That's a big debate. The thing that does seem to be settled to some extent, and we heard this a lot, was that Kevin Warsh at least is a credible person that to some extent takes the, is the Fed going to be run by somebody who's just totally-
Mirriam Gottfried: A political ideologue.
- political ideologue 政治思想に凝り固まった人物
Telis Demos: .. a political ideologue. To a lot of investors, at least that we talked to, they thought that agree or disagree with him that he was a good, solid, credible choice. He's been a Fed governor. He's been around these policy debates for a long time. So to that extent, maybe that takes some of that concern out of the equation. But I think that just because somebody's got the name and it's a person that Wall Street feels comfortable with, I don't think that we're out of the woods on the volatility from all that.
- take ~ out of the equation /ɪˈkweɪʒən/ 問題要因から除く
- out of the woods /aʊt əv ðə wʊdz/ 危機を脱して
Mirriam Gottfried: No. And in fact, the minute Kevin Warsh was named as the nominee, the markets went haywire. A lot of trades unwound. We saw gold, which we talked about last week falling a lot. Silver fell a lot. These things that had really climbed continuously, I mean, why do you think it is that those metals sold off to us?
- the minute ~ /ðə ˈmɪnɪt/ ～した瞬間に
- markets went haywire /ˈheɪˌwaɪər/ 市場が大荒れになる
haywire（ヘイワイヤー）「（機械や計画などが）狂ってしまう、おかしくなる、収拾がつかない状態になる」
もともとは「干し草（hay）を縛る針金（wire）」のこと。この針金は細くて弾力があるため、一度切れるとグニャグニャに絡まって手に負えなくなることから、「制御不能になる」 - trades unwound /ʌnˈwaʊnd/ 取引が解消・巻き戻された
Telis Demos: Well, I think it's tempting to connect that to this hawkish-dovish debate because in theory, gold and silver as commodities are inflation protection. If you think that the Fed is going to slash rates, drive up inflation, well, you would buy gold or silver to get yourself out of dollars. So maybe it speaks to the hawkish side of Kevin Warsh's views and he wants to shrink the balance sheet, which I think some people think is a tightening sort of action. Maybe that's what played out is a little bit of dialing back what people thought was going to be this huge monetary gusher from Trump-appointed leadership at the Fed.
- it's tempting to ~ /ˈtɛmp.tɪŋ/ ～したくなる
- hawkish-dovish debate /ˈhɔːkɪʃ ˈdʌvɪʃ/ タカ派・ハト派論争
- inflation protection /ɪnˈfleɪʃən prəˈtɛkʃən/ インフレ対策・ヘッジ
- slash rates /slæʃ reɪts/ 金利を大幅に引き下げる
- get yourself out of dollars /ˈdɑːlərz/ ドルから資金を逃がす
- shrink the balance sheet /ʃrɪŋk/ （FRBの）バランスシートを縮小する
- tightening sort of action /ˈtaɪtənɪŋ/ 引き締め的な行動
- monetary gusher /ˈmʌnɪˌtɛri ˈɡʌʃər/ 金融の大放出（比喩）
Gusher: 石油や水が激しく自噴する井戸のこと
短期間で莫大な富を生み出すもの、あるいは「金のなる木」
インフレ （景気過熱） 不況（景気冷え込み）
利上げ （ブレーキ） 利下げ（アクセル）
バランスシート 縮小（お金を回収） 拡大（お金を注入）
お金の流れ 絞り出す 溢れさせる
Mirriam Gottfried: But we've also seen a huge selloff in cryptocurrencies, including Bitcoin. So that's like the ultimate speculative asset. That doesn't seem necessarily consistent with what you were just saying. Maybe this goes back to what I was saying last week, which is that gold was somewhat of a momentum trade. It had become this thing that had lost contact with reality and was like a retail thing where Ma and Pa are getting out there and buying gold because they heard gold's going up.
- momentum trade /moʊˈmɛntəm/ 勢いで買われる取引
- lost contact with reality /ˈrɪˈælɪti/ 現実と乖離した
- retail thing /ˈriːteɪl/ 個人投資家主導の動き
- Ma and Pa /mɑː ænd pɑː/ 一般の普通の人たち（米口語）
Telis Demos: To your point about this speculative fervor, I mean, I think we should start talking more about crypto and what is going on in the crypto market. Obviously we know that the price of Bitcoin is down substantially, right? It was over 120,000 at one point last year. And so crypto did not participate in the gold rally, which means that the longtime view that is crypto digital gold, maybe that disproved that a little bit, but now it's falling with gold. So if gold coming down is a sign of oh, okay, people are less worried about inflation and are more comfortable with long duration assets like tech stocks are, you would see crypto rally. So now, I don't know what the heck a correlation of crypto is.
- peculative fervor /ˈspɛkjələtɪv ˈfɜːrvər/ 投機熱・熱狂
- digital gold /ˈdɪdʒɪtəl/ デジタル・ゴールド
- disproved that /dɪsˈpruːvd/ それを否定した
- long duration assets /ˈdʊreɪʃən/ 長期資産
- correlation /ˌkɔːrəˈleɪʃən/ 相関関係
- what the heck /wʌt ðə hɛk/ 一体全体（困惑・強調）
Mirriam Gottfried: I don't understand why you would ever correlate Bitcoin with gold. Personally, that doesn't make sense to me. Gold is seen as a safe asset. Bitcoin is the ultimate in speculative. It's not really tied to anything. So I think that Bitcoin is one of the riskier assets and gold should be one of the safer ones.
Telis Demos: Interesting.
Mirriam Gottfried: To me, they shouldn't trade together.
Telis Demos: Well, what I'm going to be looking at this upcoming week to start to make sense of this are a couple of earnings, one from Coinbase, one from Robinhood. These are obviously huge names in the world of crypto trading. Coinbase has a business that spans both retail and institutional crypto trading. Robinhood obviously has a big business in addition to the other things it does in options and equities and everything as we talked about, Robinhood does a lot of things these days. Crypto's a big business for them. Those stocks have come off considerably this year. They're both down more than 20% year-to-date as we're talking, matching the move in crypto. Bitcoin is down about 20. So when we say crypto, we use Bitcoin as a proxy.
- come off considerably /kəm ɔːf/ 大きく下落する
- year-to-date (YTD) /ˌjɪr tə ˈdeɪt/ 年初来
- matching the move /ˈmætʃɪŋ/ 同じ動き
- proxy /ˈprɑːksi/ 代理指標・代表例
Mirriam Gottfried: But there are a lot of other ones.
Telis Demos: Yeah. There's a lot of other crypto. So what those companies say about what they're seeing in their customers and what's going on with their volumes, I think might tell us something about what the crypto market is.
Mirriam Gottfried: Here's one thing I don't understand though. With Coinbase, I get it. Coinbase, in addition to being a trading platform, will custody the crypto assets for you. So lower values of crypto should probably correspond to lower revenues for Coinbase. But with Robinhood, isn't all trading revenue for them? Don't they just make money when people trade? So if people buy, if people sell, shouldn't they be making money? Maybe there's something about Robinhood's business that I don't understand.
- custody (the assets) /ˈkʌstədi/ （資産を）保管・管理する
- correspond to /ˌkɔːrəˈspɑːnd/ ～に対応する・連動する
Telis Demos: No, no. I think that's right to a degree. I don't think that we necessarily know the one for one relationship between the price of Bitcoin and how much trading happens. I think historically, and they've talked about this in the past, lower price means lower volume. Essentially, just the air comes out of the crypto room. Maybe that relationship is breaking down. We should see the latest evidence of that. Also, when the price of Bitcoin comes down and it falls so substantially, I mean, we're talking almost half of what it was at one point last year. That means that even if you are buying and selling a lot, just the amount of dollars, the notional volume of your trade has shrunk. And these guys make money, not just from every time you trade, but the size of the trade in dollar terms, a percentage of that.
Mirriam Gottfried: I see. A percentage of that.
- one-for-one relationship /ˌwʌn fər ˈwʌn/ 一対一の関係
- the air comes out /ɛr/ 熱気が抜ける（比喩）
- notional volume /ˈnoʊʃənəl/ 名目取引額
- in dollar terms /ˈdɑːlər/ ドル換算で
- a percentage of that /pərˈsɛntɪdʒ/ その一定割合
Telis Demos: So there's some element to that. It's now, obviously if-
Mirriam Gottfried: So if crypto's one of the biggest things traded on Robinhood and it's worth less, and most of these coins are worth less then, yeah.
Telis Demos: You look at other brokerage stocks, Charles Schwab, their stock is up this year. Interactive Brokers, which is even more active trading related, their stock is up this year as well. So it doesn't seem like there's this washout in investors belief and brokerage and speculative activity. I think it's a crypto thing.
- brokerage stocks /ˈbroʊkərɪdʒ/ 証券会社株
- even more active trading related /ˈæktɪv/ 取引量への依存度がさらに高い
- washout /ˈwɑːʃaʊt/ 総崩れ・全面的な崩壊
Mirriam Gottfried: And maybe it's also just this overall momentum thing. Maybe people are starting to say, "Okay, things have gotten too heated in a variety of categories," and a lot of it's been driven by the retail customers who use Robinhood.
Telis Demos: Well, and interestingly too, and this will connect to our next big topic of what's going on in software. If investors are worried about crypto, if they're not worried about people in general, okay, people will still buy and sell stocks. People aren't going to pack up and put cash under the mattress. So they like stocks, Charles Schwab or Interactive Brokers, but they don't like Robinhood right now. They also don't like the other poll, which is like alternative assets. The retail push into things like private credit and private equity, and you've seen a big sell off.
- the other pole /poʊl/ もう一方の極（対極）
- alternative assets /ɔːlˈtɜːrnətɪv/ 代替資産
- retail push into /ˈriːteɪl/ 個人投資家の流入
- private credit /ˈpraɪvət/ プライベートクレジット（非公開融資）
- private equity /ˈɛkwɪti/ プライベートエクイティ
- big sell-off /ˈsɛl ɔːf/ 大きな売り
Mirriam Gottfried: People are getting really nervous about that. And there are a variety of reasons behind that, some which are grounded in fact, and some which might just be more in sentiment.
- grounded in fact /ˈɡraʊndɪd/ 事実に基づいた
- sentiment /ˈsɛntɪmənt/ 市場心理・雰囲気
--------------------------------------------------------------------------------------FEBRUARY 8b 2026
9：42
Telis Demos: Tell us about what happened in software this past week and how it relates to what's going on with those alternative asset managers.
Mirriam Gottfried: So we saw big selloffs in the Nasdaq driven by software stocks. Anthropic released a new product or announced a new product. I don't even know how available it is at this point, but basically it's an AI tool for legal purposes. And people saw that as being a big threat to software stocks saying, "Oh, maybe companies won't buy software as a service. These big software companies that have made so much money over the past few years and have become such a popular trade because they'll just be able to create what they need with an AI tool." And so that drove not only big software stock selloffs, but also selloffs in lenders to software companies that sit in private equity portfolios. So a lot of private equity firms have bought software companies because of this recurring revenue, the cash flow that they generate.
- software as a service (SaaS) /sǽs/ 定額制クラウドソフト
- recurring revenue /rɪˈkɜːrɪŋ/ 継続収益
- cash flow /ˈkæʃ floʊ/ キャッシュフロー
- per seat /pər siːt/ ユーザー単位で
Telis Demos: Because the whole model of cloud is, it's not just like the old days when, okay, you want Lotus 1-2-3, you go buy a copy of Lotus. You walk into, whatever.
Mirriam Gottfried: You stick the floppy disk in.
Telis Demos: We walk into Babbages or whatever, you buy the box, you take it home, you stick. Right, now of course, we get our software over the internet and you pay a regular fee like you do for Netflix or something.
Mirriam Gottfried: And it's per seat, right? So if you're a worker at a company, the company pays for your license and every month they pay for that.
Telis Demos: And the theory was it's not like Netflix where you can cancel it at any time. It's you subscribe to these, you have contracts, you're tied into them for years. It's impossible to change, right?
Mirriam Gottfried: And your entire architecture for your company is in many cases built on this. So this made it a very popular play, not only for public markets, but also for private markets. And a lot of the loans to those private market companies were made by private credit providers like Blue Owl, Aries, et cetera. And so those stocks sold off a lot too. And the BDCs, the business development companies, which are basically direct lenders, lenders to these kinds of companies sold off. But I think a little bit of that, and maybe some of the software sell off itself, probably a good amount of it was overblown because it's been pretty indiscriminate. Not every piece of software is going to be disrupted by AI and it's not going to happen overnight if it does.
- popular play /ˈpɑːpjələr pleɪ/ 人気の投資対象
- BDC (Business Development Company) /ˌbiː diː ˈsiː/ 中堅・未公開企業向け直接融資会社
- indiscriminate /ˌɪndɪˈskrɪmɪnət/ 見境のない、無差別な
- overblown /ˌoʊvərˈbloʊn/ 誇張された、行き過ぎた
Telis Demos: Yeah. Our hard colleague, Dan Gallagher, who is really an expert in this stuff, he wrote after that big software sell off, he said, "Look, it takes a long time to replace these things." "And even if there is some disruption to them from, 'Oh, okay, I can do the same thing by just asking Claude to do it for me,'" right, an Anthropics AI tool or ChatGPT or whatever, "I don't need to pay for this software anymore." Well, but those companies are also adding AI into how their software operates. And some of the companies we're talking about here are the sales forces of the world, like companies that sell you day-to-day business software.
Mirriam Gottfried: The companies aren't going to just throw out software that they use to manage very sensitive customer relationships overnight. It would be very risky at this point to say, "Okay, we're going to replace that with AI."
Telis Demos: Well, and remember the big DeepSeek catastrophe for AI last year, right?
Mirriam Gottfried: Which happened almost about a year ago.
Telis Demos: And where did AI stocks go from there?
Mirriam Gottfried: Up, right. So there was a big scare because of a new AI related technology and stock sold off. And if you bought then, you would have made money. So I think there could be a buying opportunity right now in software for those who are willing to stomach the negative headlines for a little bit. Plus, I think that some of these lenders have also sold off too much. Apollo, for example, has been out there saying they already sold off a lot of their software exposure. They see software as a risk and they're not very exposed on the equity side or on the debt side to software.
- buying opportunity /ˈbaɪɪŋ ˌɑːpərˈtuːnɪti/ 買い場
- stomach (headlines) /ˈstʌmək/ （嫌な状況を）耐える
Telis Demos: Interesting. Timely, timely. Apollo's earnings report is coming up this upcoming week. That's one I think to watch closely. There's also TSLX, which is a BBC backed by Sixth Street, again, a giant private asset investor. That's another one, I think they often have a very interesting investor letter that addresses big topics in private credit. That'll be a very interesting one to watch as well.
Mirriam Gottfried: I agree.
Telis Demos: Well, I want to highlight actually... So we've been in a conversation with our audience about how they invest. And as we've discussed, a lot of people are buy and hold and like to think a little longer term. And I look back and we actually had an interesting thought from one audience member, Jed Carroll, he wrote in to talk about things that he's thinking about. And one thing was that he was already on top of the idea. He said, "Machine learning is driving a debt bubble in the US. AI won't go away, but we could have something like the telecom bubble." So some of our listeners, I think, were already thinking about risks here, but also, but I asked Jed, I said, "Hey, by the way, do you have any reactions to what's happened most recently in software?" And he said something interesting. He said he thought that the software runs are irrational, that successful AI implementation is helped by structured data. And that's one thing I thought that was a really interesting comment because I've heard the same thing from some lenders to software companies talking about how, "Look, one reason that we invested in this particular company is because they're not just some piece of software that does something that sure could be automated." "It's they also manage your data and hold your data and that's proprietary data that you can't just wipe that away. You'd have to rethink your whole architecture of your company." So I think it's interesting that the Jed is thinking about this. Like, "All right, yeah, there's bubbly stuff, but also some of this stuff might be overstated." So always interesting to see how our audience is thinking about this stuff.
- top of the idea /tɑp/ その考えを理解・把握している
- runs are irrational /ˈɪˌræʃənəl/ （株式の）上昇や売りが非合理的である
run : 株価や資金の急激な上昇・下降を指す - structured data /ˈstrʌktʃərd ˈdeɪtə/ 構造化されたデータ（表形式など、AIに扱いやすいデータ）
- proprietary data /prəˈpraɪətɛri/ 独自データ、企業秘密データ
- wipe that away /waɪp/ 消す、なくす
- overstated /ˌoʊvərˈsteɪtɪd/ 過剰に強調された、誇張された
- architecture of your company /ˈɑːrkɪˌtɛktʃər/ 会社のシステム全体・組織構造
Mirriam Gottfried: And speaking of a bubble, Hemant Taneja, who I spoke with for WSJ Invest Live, which is coming up next, addressed the issue of a bubble, sort of. He got into it. I asked him about it.
Telis Demos: Well, he was the perfect person to talk to about this, right?
Mirriam Gottfried: And he's an investor in Anthropic.
Telis Demos: There you go.
Mirriam Gottfried: The thing that set this whole software sell off in motion.
Telis Demos: And for listeners who don't know, I mean, he's a legendary venture capital investor, right? And he's the CEO of General Catalyst and they have backed companies like Airbnb, Stripe, Snap. I mean, some of the biggest venture successes. So he's been doing this for a long time and you talked to him about how and when and why he got into AI and he wasn't necessarily doing it right at the very beginning either.
Mirriam Gottfried: No, it actually took him a while to be convinced and we'll hear more about why that is coming up in a bit.
Hello. I'm going to dive right in. So the Wall Street Journal has reported that Anthropic, which you invested in and OpenAI are in a race to go public this year at huge valuations. Meanwhile, JPMorgan is estimating 5.3 trillion in spending on AI infrastructure and data centers over the next five years. Are we in a bubble?
Hemant Taneja: Oh boy.
Mirriam Gottfried: Starting off with a big one.
Hemant Taneja: Starting off with the big one. So look, I would say it's very nuanced. And I always say this, bubbles are good because they actually mobilize talent and capital into important areas. So let's not be afraid of bubbles. But I will say it's more nuanced because on one side you're seeing real application of AI and how it's transforming labor into productivity. I think that's a huge economic opportunity. We can talk about the complexities of it, but on the other side, there is a lot of science project stuff being funded around, is there a better model out there that's going to beat OpenAI and Anthropic and Gemini? And so there's a ton of capital going into the venture capital industry around things that are unclear what will happen to them, but then there's also a lot more where the bigger economic opportunities that is practical. So it's not quite simple an answer there.
Mirriam Gottfried: I don't know if you fully answered that question, but maybe we'll get more to it later in the interview. So you first invested in Anthropic last year at, I think, a $61.5 billion valuation. Why did you wait so long to invest in an LLM and has your view of that business model evolved?
Hemant Taneja: I'll tell you, so I've been investing for 25 years in venture, and this has just been the hardest time because these models take a lot of capital. And so how much capital would you put at risk before you know there's a practical application? So when we invested at $60 billion valuation, the business was doing approximately about $880 million of run rate on a revenue basis. And we modeled it to be, okay, we'll probably do two to three billion, which was last year, and now it's starting to look like there's a real business. We're still paying 18 months ahead, but we can start to see what these things are going to become in terms of real applications. There was a cloud code application emerging. You could see that this is not just the promise of AI, but other real good economics that you could get behind. And I think the market just significantly changed. We did not invest in these things before that because we couldn't get our arms around.
- run rate /rʌn reɪt/ 年間換算売上（現状の売上ペースを基に年間換算）
- cloud code application /klaʊd koʊd ˌæplɪˈkeɪʃən/ クラウド上で動作するコード/アプリケーション
- real applications /rɪəl ˌæplɪˈkeɪʃənz/ 実用的な応用例
- get our arms around /ɡɛt aʊər ɑrmz əˈraʊnd/ 理解する、把握する
Mirriam Gottfried: Really interesting. Are you worried about the argument that this was brought up last night that Yann LeCun has been making, that the current big LLMs have reached a dead end and can't get smarter?
Hemant Taneja: Look Yann's been saying that for a while. I think there are two things happening. One is scaling laws have continued to work and that's actually the fundamental belief as to why there's so much infrastructure spend going in because with more capital, more compute, more energy, you could continue to scale these models. So I do think that hasn't really changed, but intuitively, I do think language is a constraining view of the world. And you think about the fact that as human beings, we understood and appreciated the universe and put it in language, and these models are based off of that. And Yann's thesis around, well, there's this world model which is observe the world in a different way than being constrained by language, and could you create a next level of intelligence? Theoretically, that's possible. And I do think there are good teams working on that attempt, but that does not mean that the potential for language models has tapped out.
- scaling laws /ˈskeɪlɪŋ lɔːz/ スケーリング則（モデルの性能やデータ量・計算資源の関係法則）
- infrastructure spend /ˈɪnfrəˌstrʌktʃər spɛnd/ インフラ投資
- compute /kəmˈpjuːt/ 計算能力、コンピューティングリソース
- intuitively /ɪnˈtuːɪtɪvli/ 直感的に、感覚的に
- constraining view of the world /kənˈstreɪnɪŋ vjuː əv ðə wɜːrld/ 世界の見方に制約を与えるもの
- world model /wɜːrld ˈmɑːdl/ 世界モデル（AIが世界を理解する内部表現）
- tapped out /tæpt aʊt/ 限界に達した、使い切った
Mirriam Gottfried: So on a recent episode of WSJ's Take On the Week podcast, we discussed how the hyperscalers say that there will be only one winner in this AI race. And do you think that's true? And what about with the LLMs? Do you think there will be only one winner? Is this a winner take all market?
- hyperscalers /ˈhaɪpərskeɪlərz/ 大規模クラウド事業者（AWS, Google Cloud, Azure など）
- LLMs /ˌɛl ɛl ˈɛm/ Large Language Models（大規模言語モデル）
- winner take all /ˈwɪnər teɪk ɔːl/ 勝者総取り（1社だけが市場を独占すること）
Hemant Taneja: See, I don't subscribe to that. So if you think about the three models that have scaled, you've got Anthropic OpenAI, you've got Gemini. You would say Anthropic... Every one of them has a killer app that's actually driving the use of the models. Obviously there's a general intelligence that they're building, but if you go understand what Anthropic's doing with Claude, it truly is about how does software get written and what is our relationship with software as we digitize our businesses. It's a killer app. It's working really well. I mean, it's becoming more and more powerful. If you look at OpenAI, yeah, they talk about an enterprise business, but a lot of the attraction is really is ChatGPT. It's a ubiquitous application, changed all of our lives. We all sound a lot smarter in our cocktail conversations because of it. And I think there's a business there. And Gemini, if you look at Google, it's actually more in the consumer end as well. So they really are becoming app companies that have models underneath. The horizontal opportunity of, is there going to be one model that takes on, it's hard to believe? I really do think they'll specialize in different areas.
- subscribe to /səbˈskraɪb tu/ 信じる、支持する（ここでは「～には賛同しない」）
- killer app /ˈkɪlər æp/ 特定の技術や製品の成功を牽引するアプリ・サービス
- ubiquitous /juːˈbɪkwɪtəs/ 至る所にある、普及している
- horizontal opportunity /ˌhɔːrɪˈzɑːntl ˌɑːpərˈtuːnɪti/ 横方向の機会（業界全体で一つのモデルが独占できるかという意味）
Mirriam Gottfried: So many companies are talking about integrating AI into their operations right now, and everyone's investing a huge amount, but so far it seems like the gains from that investment haven't quite showed up in a lot of different areas. Where have you seen it most successfully integrated and why do you think that was success?
Hemant Taneja: So I think, first thing, when we talk about enterprise diffusion of AI, you have to understand what are all the things that have to happen for this to really work. So from our standpoint, you need your data infrastructure to be ready, which is a lot of the draft behind Databricks and companies like that. You need the models to be adapted to your language, your secret sauce, your understanding of the business. And then you have to think about this immense workforce transformation because you're going to have human beings and AI agents all working together. Some humans are managing agents, some agents are managing humans. There's a fundamentally different organizational structure of a company that's going to get run. What does culture really mean? All those things have to be rethought. So when you look at all that and you say all this really started three years ago, this idea that this was going to diffuse so fast, to me, it was a bit of a fool's errand. So it does take some time. Now, we fund a lot of companies and we build a lot of companies that are taking on this problem and I look at their progress and the pace of progress, it's actually pretty stunning. And we will look in five years and say the labor productivity mix is fundamentally changing. And I'll give you an example of that. We launched a company called Hippocratic AI. This was about two and a half years ago. We got a lot of health systems to made over 10 million phone calls, fully automated in hospitals, checking on patients, either the ones that had a procedure the day before or the ones that are about to come in to make sure they're ready. That would have cost $90 an hour with a nurse and we don't have enough nurses. And that's now being done at a 10th of the cost and you can actually take care of so many more patients as a result. So 10 million phone calls will become 10 billion phone calls just in terms of just scaling compute capacity and the organizations saying they're going to go deeper. So you're starting to see examples like this where there truly is the potential of fast diffusion. And I've said this before, last year was a year diffusion truly started to take hold. We started to understand how it's really going to work and it will be on a relatively accelerated path.
- enterprise diffusion /ˈɛntərpraɪz dɪˈfjuːʒən/ 企業への浸透・導入
- data infrastructure /ˈdeɪtə ˌɪnfrəˈstrʌktʃər/ データ基盤・インフラ
- secret sauce /ˈsiːkrət sɔːs/ 独自ノウハウ、競争優位の秘訣
- workforce transformation /ˈwɜːrkfɔːrs ˌtrænsfərˈmeɪʃən/ 労働力の変革
- scale compute capacity /skeɪl kəmˈpjuːt kəˈpæsɪti/ 計算リソースを拡大する
- fool's errand /fuːlz ˈɛrənd/ 無謀な試み、無駄骨
- relatively accelerated path /ˈrɛlətɪvli əkˈsɛləreɪtɪd pæθ/ 比較的加速した道筋
- get run → 「実行される／運用される／稼働される」
Mirriam Gottfried: It seems like the reward for successfully integrating AI is a lot of job loss. So for the average person, how is AI actually benefiting them in the five?
Hemant Taneja: Yeah. So I think first thing is, this is where the trickiness of adoption of AI comes in, that there's a fourth element to how companies have to adopt, which is the courage of the leadership. Because if you bring AI in, there's a fundamental fear that, "What happens to my jobs?" And what I advise, we do a lot of work with organizations on their AI transformation. I always say, "Don't focus on trying to drive efficiency with AI, focus on your growth." And what that does is your people have more interesting work to do, they grow, they are excited about winning together as opposed to efficiency means there's going to be a job loss. So the first thing is culturally, how do you really transition yourself to be an AI person-
- integrating AI /ˈɪntɪˌɡreɪtɪŋ eɪˈaɪ/ AIを統合する、導入して機能させる
- reward /rɪˈwɔːrd/ 報酬、リターン、成果
- the trickiness of adoption /ðə ˈtrɪkɪnəs ʌv əˈdɑːpʃən/ 導入の難しさ、複雑さ
- courage of the leadership /ˈkʌrɪdʒ ʌv ðə ˈliːdərʃɪp/ 経営陣の勇気、意思決定の勇気
- fundamental fear /ˌfʌndəˈmɛntl fɪr/ 根本的な恐怖、不安
- drive efficiency /draɪv ɪˈfɪʃənsi/ 効率性を高める
- focus on growth /ˈfoʊkəs ɑn ɡroʊθ/ 成長に注力する
- culturally transition /ˈkʌltʃərəli trænˈzɪʃən/ 文化的に適応・移行する
Mirriam Gottfried: So you're saying there shouldn't be job loss?
Hemant Taneja: Well, no, I'm saying, so I think in terms of being adopted, that's the case. Job loss is a real thing. I think in the short term, we actually need more jobs. We need people to learn how to train. We actually need education and curriculum to change so we're better at taking advantage of these prompting technologies in our day-to-day work. I think medium term, there's new kinds of jobs that we can envision. So if you go back to the nurse example I gave, if it's effectively free to check up on our elderly, you would fundamentally rethink the care model for a part of healthcare that's almost half the cost. We just take care of folks that are above 65, 70 in the US healthcare system. There's a whole infrastructure that would get created around that to go drive that abundance. Long term, I do think there's a real issue around what happens with jobs because, and I do think it's different this time than all the examples people give from before because never ever was there a case that every skill in any job can be better done by AI and robotics, which is I think where we're headed in the next not five years, but 20 to 30 years. So we do have a long-term issue that we have to address, but in the short term, I do think it creates opportunity as long as we're focused on re-skilling people to take advantage of AI effectively.
- prompting technologies /ˈprɑːmptɪŋ tɛkˈnɑlədʒiz/ AIに指示を与えるプロンプト技術（入力指示を活用する技術）
- re-skill / re-skilling /ˌriːˈskɪl/ 再教育、スキルの再習得
- abundance /əˈbʌndəns/ 豊かさ、十分な資源や機会
- fundamentally rethink /ˌfʌndəˈmɛntəli riːˈθɪŋk/ 根本的に再考する
Mirriam Gottfried: Do you think that governments should play a role in preparing for that long-term issue?
Hemant Taneja: We all need to. I think it's a responsibility of businesses. It's a responsibility of government. It's not going to happen unless there's true public-private partnership on thinking about changing the education curriculum. It's thinking about changing and being accelerated about re-skilling people into new skills. This is not the thing that we should wait 10 years and figure how to deal with. We should be working on this now.
Mirriam Gottfried: I want to switch gears a little bit. Last week, you were one of the few CEOs to make a public statement following the shooting of a protestor by an ICE agent in Minneapolis. You wrote on X that what we were seeing was quote, "A threat to the promise of America." Why do you think that other leaders aren't speaking out? There must be others who agree with you.
Hemant Taneja: Yeah. Look, we're in a really interesting time because everybody that's starting to create a voice optimizes for what do I do on social media? And by definition, it makes you go to the fringes. And the crazier the positions you take, the more you get a voice and then you rationalize to yourself, "Well, I have voice now and I can do what I think is the right thing." And you decouple how you communicate from what your actions are. And I think that's a broken phenomenon. And so people take these radical views on both sides and we have, as a business explicitly said, "We're just going to be apolitical. We're going to focus on policy. We're going to focus on role of business in society and also figure out how to build businesses that drive opportunity for everybody." That's actually fundamentally what this election was all about. It really was about creating opportunity for everybody. But we deviate from that and we don't speak up on moments like this where these things are blatantly wrong. I do think that's a lost opportunity for business leaders to be leaders in society and we want to take a position on that.
- optimize for /ˈɑptɪmaɪz fɔr/ ～に最適化する、～を最大限活かす
- fringe / fringes /frɪndʒ/ 過激派、周縁（主流から外れた立場）
- decouple /diːˈkʌpəl/ 切り離す、分離する
- apolitical /ˌeɪpəˈlɪtɪkəl/ 政治に関わらない、中立的な
- blatantly wrong /ˈbleɪtəntli rɔŋ/ 明らかに間違っている
Mirriam Gottfried: So speaking of building a business, General Catalyst has launched a number of new strategies, including a private wealth business, which may have led to the speculation last year that you were planning to go public. Are you planning to go public?
Hemant Taneja: No.
Mirriam Gottfried: Okay. So if you're not planning to go public, could you tell us a little bit more what the rationale was behind these new businesses?
Hemant Taneja: Yeah, look, I think, so we've been around for 25 years. We've backed companies like Kayak and Stripe and Airbnb and Snap and Anduril and Anthropic and others. And our business at the end of the day is all about helping great founders build enduring companies. So everything we do, one lens to look at it is, are we meeting founders where they are and are we giving them unfair advantages to win? And so everything we've done has been in that cost. We care deeply about relationships. We call it our familia. And part of having a deep relationship with our founder community and our broader ecosystem is help them deal with their wealth issues. So the goal is not to build the biggest wealth management business, it's really to serve them. We got into the healthcare business that was all about how do we actually create conditions for healthcare founders to figure out how to solve problems at scale?
- enduring companies /ɪnˈdjʊərɪŋ ˈkʌmpəniz/ 長期的に存続する会社、持続可能な企業
- one lens to look at it /wʌn lɛnz tu lʊk æt ɪt/ 一つの視点で見ると
- meeting founders where they are /ˈmiːtɪŋ ˈfaʊndərz wɛr ðeɪ ɑr/ 創業者の立場や状況に合わせて支援する
- unfair advantages /ʌnˈfɛr ˈædvəntɪdʒɪz/ 他者に対して有利な条件（競争優位性）
- familia /fəˈmiːliə/ 家族的なつながり（ここではコミュニティの意）
- wealth issues /wɛlθ ˈɪʃuːz/ 資産や財務上の問題
- serve them /sɜrv ðɛm/ 支援する、役に立つ
- at scale /æt skeɪl/ 大規模で、広範囲にわたって
Mirriam Gottfried: You actually bought a hospital.
Hemant Taneja: We bought a hospital. I have seven of our companies that I'm actively working with designing what is an AI native hospital going to look like and go do it at the hospital we bought. We didn't buy it as an investment out of our funds. We bought it off our balance sheet actually to create conditions and create a model that other institutions across the country can actually be adopting as well. And so all of our businesses that you look at are essentially transformation infrastructure that our founders can plug into because at the end of the day, AI is a transformative opportunity. The biggest value is going to accrue to companies that have the courage to transform because they have the data, they have the customers, they have the scale, and that's where AI can do its magic.
- AI native hospital /eɪ aɪ ˈneɪtɪv ˈhɑːspɪtl/ AIを前提に設計された病院
- balance sheet /ˈbæl.əns ʃiːt/ 財務諸表、企業自身の資産を使った購入
- plug into /plʌɡ ˈɪntu/ 接続する、活用する、取り入れる
- accrue to /əˈkruː tu/ （利益・価値が）…に集まる、帰属する
- courage to transform /ˈkʌrɪdʒ tu trænsˈfɔrm/ 変革する勇気
- scale /skeɪl/ 規模（事業規模、顧客基盤など）
- AI can do its magic /eɪ aɪ kæn du ɪts ˈmædʒɪk/ AIが最大限の効果を発揮する
Mirriam Gottfried: I think we're out of time. Thank you so much.
Hemant Taneja: Thanks for having me.
Telis Demos: And that's everything you need to know to take on your week. This show is produced by Anthony Bansie and Michael LaValle. Michael LaValle and Jessica Fenton are our sound designers. Michael also wrote our theme music. Jessica Fenton was our technical manager. Aisha Al-Muslim is our development producer. Chris Zinsli is our deputy editor, and Philana Patterson is the head of news audio for The Wall Street Journal. For even more, head to wsj.com. I'm Telis Demos.
Mirriam Gottfried: And I'm Mirriam Gottfried. Until next time.
Telis Demos: I'm glad I went to the mirror before this because I had a little pen explosion issue.
Mirriam Gottfried: You had pen on your face.
Telis Demos: And apparently I had absentmindedly done this and I just had a big blue streak.
- absentmindedly /ˌæbsəntˈmaɪndɪdli/ ぼんやりして、うっかりと