Title: Google's Negative Cash Flow and the AI Capex Reckoning | The Weekly Wrap Show: The Real Eisman Playbook — The Weekly Wrap Guest: Steve Eisman (host, solo episode) Date: 2026-07-24 (recorded Thursday night, July 23; "week ending Friday, July 24") URL: https://youtu.be/983SbhP0aRs Length: ~21:45 Note: Scripted monologue — already close to clean. Remove-only cleanup per skill Step 1: the [music] / [snorts] audio artifacts and the ">>" speaker-change tics were deleted; no words were changed, added, reordered or paraphrased. Auto-transcript name garbles are LEFT INTACT in this raw file and corrected only in the analysis pages (e.g. "Ken Sahawski"/"Ken Sahousky"→Ken Suchoski of Autonomous Research, "Brad Safalow"→Brad Safalow of PAA Research, "SASpocalypse"/"cesspool narrative"→SaaSpocalypse, "Ben Kallo"→Ben Kallo of Baird, "Gil Luria"→Gil Luria of D.A. Davidson). Every (mm:ss) cue preserved exactly where it was.
00:00 The big news was that a Chinese AI company announced the release of its new LLM model but for a fraction of the cost. Now, the possibility of a price war looms closer. Domino's is a poster child for the K-shaped economy. It's down 20% this year. A bunch of very important companies reported. Google, Tesla, ServiceNow, and IBM.
00:20 Let's take them one at a time. A year ago, it was all rah-rah for AI. As the AI story has somewhat matured, the story has shifted. China AI players have produced great models that are much cheaper and might create a price war. Everyone is nervous and that nervousness was on full display this week. Hi, this is Steve Eisman and this is another episode of the weekly wrap.
00:51 This is for the week ending Friday, July 24th, but recorded Thursday night, July 23rd. Before we get to the wrap, I would like to remind everyone about our move to Substack and explain the value add. Substack is an exciting community of like-minded investors and the conversations are dynamic. The Substack ecosystem is well-established with a large variety of podcasters that I interact with regularly.
01:17 As a free Substack subscriber, you will receive emails sent directly to you every time we release a new premium episode and a sneak peek preview of both the video and newsletter. You'll also have access to our notes and restacks. The link to join for free is in the description. Let me quickly flag what is in our premium Substack subscription of late.
01:38 On Wednesday, July 22nd, we released an interview with recurring guest Ken Sahawski, the payments analyst at Autonomous. We discuss how AI and agentic AI are changing the payments landscape. And on Wednesday, July 29th, we will release an episode with recurring guest Brad Safalow. Brad has a specialty in providing research where he recommends shorting certain companies.
02:01 He also has a vast expertise in the real estate brokerage business. We discussed how the real estate brokerage ecosystem is changing, and we also delved into some of his shorts, including companies in the for-profit education industry. The link for premium is in the description. Before I get to the wrap, let me point out that Charter is reporting this Friday, but because we record the wrap Thursday night, I will be commenting on it next week.
02:26 On this week's wrap, we will discuss the war in Iran, AI developments, a whole bunch of companies reported, investors are showing their displeasure with ever-mounting AI CapEx, and two mailbags. Let's get started. Over the weekend, the US and Iran traded blows, and it was reported that several US soldiers had been killed.
02:46 Things seem to be escalating. Later in the week, the Houthis decided to get involved and bombed some Saudi tankers. As a result, oil prices climbed to 100, and the yield on the 10-year reached 4.7%. President Trump threatened more attacks on Iran. Last week, the big news was that a Chinese AI company, Moonshot, announced the release of its new LLM model called Kimi K3.
03:10 Moonshot claimed that Kimi K3 is as good as any LLM out there, but for a fraction of the cost. Prior to last week, we were worried about AI capital intensity and the lack of moats. Now, the possibility of a price war looms closer. Moving on. SpaceX is now well below its IPO price. I'm not sure what this means yet, but it does not bode well for the IPO market.
03:32 And let's get to companies that have reported. First up, Domino's Pizza reported, and the stock was up a bit on Monday on the print. Domino's is a poster child for the K-shaped economy. It's down 20% this year. In the March quarter, EPS was down 5%, but in this quarter, EPS was up 7%. However, EPS missed expectations, but revenue beat, and the revenue beat caused the stock to climb 2% on Monday.
03:59 Don't get carried away. Domino's same-store sales growth, which is the best indication of growth for a consumer-facing company, fell to its lowest pace in five quarters, a mere 0.1%. After Monday, Domino's gave back all of its gains plus. Now, normally companies don't report on Monday. Domino's is an exception.
04:20 Tuesday witnessed more reports. Equifax reported. Now, we have not really spoken about Equifax before, except in the context of my short thesis on FICO. Equifax is one of the three credit bureaus. Now, while all three credit bureaus provide consumer information for scoring purposes, they also have different business mixes. On the scoring side, Equifax is heavily mortgage-dependent, but Equifax's largest business is not scoring nor scoring-related.
04:51 It's largest business is called Workforce Solutions division, EWS. And that is a data and technology business that provides automated verification of income and employment. EWS provides this service to businesses and to federal, state, and local governments. In other words, EWS is a software business. So, Equifax's EWS has been part of the SASpocalypse debate.
05:16 Bears have been arguing that EWS is bound to lose share to AI-powered verification services. Now, because of the SASpocalypse, there is no room for error. Unfortunately for Equifax, Q2 government revenue growth was down mid-single digits, and was below both management guidance of flat year-over-year and below street expectations.
05:40 Third quarter EPS guidance is 3% below consensus, and the implied Q4 EPS guidance is also 3% below street estimates. The lesson here is that Equifax management might have a legitimate reason for the weakness in government EWS revenue. And the company blamed state government budget concerns. But in an environment where the cesspool narrative still reigns supreme, no one is interested in excuses.
06:08 Prior to Tuesday, Equifax was down 17% this year and down 30% over the last year. On this news, the stock was down an additional 4%. Moving on. It's been a tough period for most auto companies, but General Motors has been executing well. The company reported earnings per share of 357 versus 253, which is 41% growth, which is impressive.
06:30 And that's versus expectations of 319. So a big beat. It raised full-year profit guidance and the stock was up on the news. However, not all is great. Despite the raising of earnings guidance, US sales fell year-over-year, including sales of large pickup trucks and SUVs, which make up most of GM's earnings. Moving on. Given the geopolitical situation, it is unsurprising that defense companies are doing well. Northrop Grumman reported.
06:57 The company reported earnings per share of 768 versus 815 and versus expectations of 682. Revenue beat as well. The backlog increased by 20 billion to reach a record of 105 billion and the company raised EPS guidance. But the stock was down on the open anyway, then it recovered. But it was not up on these good results.
07:19 Why? The cost growth on the company's missile programs seem open-ended and that is hurting current margins. Lockheed Martin also reported and reported great numbers. Lockheed posted very strong 2Q 26 results with sales up 11% and earnings per share of 794 versus a dollar 46 last year and versus 720 expected. Revenue beat as well.
07:43 The backlog reached a record of 230 billion, up 24% in 3 months. Clearly, the geopolitical situation is benefiting defense companies. Moving on. GE Vernova reported Wednesday morning. The results were great, but the stock went down anyway. First, the facts. GEV is one of the best AI-related power stories. Its power division produces gas turbines for utilities.
08:07 There are only three companies in the world that produce large gas turbines: GE, Mitsubishi, and Siemens. Its electrification division manufactures all kinds of equipment used by utilities and other power producers. GEV got spun out of GE in April 2024 at $143. Because it's one of the best AI power stories, the stock has climbed to over $1,000.
08:33 This quarter, EPS was 247, 33% growth, but a miss versus expectations. Revenue of 11.1 billion was 22% higher than last year. More importantly, orders of 24.2 billion were up 88% versus last year, and that lifted the backlog to 176 billion. The company raised revenue and EPS guidance, but the raise was below some of the whisper numbers out there.
08:55 That's why the stock was down on the print. I don't think that most investors are focused on the EPS miss. This is a very long-tail business. The most important metric is orders, and that was up 88%. I still own the stock, and even at these nosebleed elevations, I remain confident in this investment. Moving on.
09:13 Moody's reported. I've owned this stock for years because it is a duopoly with pricing power. Moody's reported a very good quarter. Earnings per share was 468, up 31% versus last year. Now, Moody's is down a bit this year as some investors have assumed that AI could eat into the Moody's and S&P duopoly.
09:32 I do not believe that is possible, so I continue to be an owner. The stock has been flat since the fall of 2024, so the multiple has come down. The 2026 PE multiple is now sub 30 times, which is the cheapest the stock has been in quite some time. Wednesday night, wow, Wednesday night was a big night. A bunch of very important companies reported.
09:53 Google, Tesla, ServiceNow, and IBM. Let's take them one at a time. Hi, Steve Eisman here. Hiring people can be really difficult. Finding candidates, sorting through resumes, getting in touch with candidates, it's all really time-consuming. Well, the future of hiring looks much brighter because ZipRecruiter's latest tools and features help speed up finding the right people for your role, so you save valuable time.
10:19 And now, you can try ZipRecruiter for free at ziprecruiter.com/eisman. With ZipRecruiter's new advances, you can easily find and connect with qualified candidates in minutes. See a candidate you are really interested in? You can unlock their contact info instantly. Over 320,000 new resumes are added to ZipRecruiter monthly, which means you can reach more potential hires and fill roles sooner.
10:41 No wonder ZipRecruiter is the number one-rated hiring site based on G2. Use ZipRecruiter and save time hiring. Four out of five employers who post on ZipRecruiter get a quality candidate within the first day. And if you go to ziprecruiter.com/eisman right now, you can try it for free. Again, that's ziprecruiter.com/eisman.
11:01 ZipRecruiter, the smartest way to hire. Google. In my view, these were very mixed results at best. Google had massive gains on its investments, which it reported as part of earnings per share. So, I take that out. Adjusted EPS was 285 versus 231 last year, but a miss versus expectations of 289.
11:23 But, revenue increased an impressive 24%. Google Cloud revenue reached 24.8 billion, up an incredible 82%. So, so far, not bad. On the bad side, however, because of the massive AI CapEx spending, free cash flow turned negative. It was negative 5.9 billion. Moreover, Google upped its 2026 AI CapEx spend from 190 billion to 205 billion.
11:54 That's a lot of billion, and the market is beginning to lose patience with all this crazy spending. The stock was down after hours. Tesla, very mixed results. EPS was 33 cents versus 40 cents last year, and versus 51 cents expected, so a big miss. Revenue was good at 28 billion versus expectations of 26 billion. The problem here is margins.
12:18 They were under pressure partially because of a 67% decline in regulatory credits, which the Trump administration has largely eliminated. Also, and maybe even more importantly, Tesla grew its CapEx spending to 5.8 billion, thereby sending its free cash flow into the red for the first time in 2 years, despite the nice growth in revenue.
12:39 Free cash flow was a negative 1.1 billion. Like Google, Tesla was down after hours. IBM, we spoke about IBM last week because IBM negatively pre-announced, and the stock was down last week on that day 25%. The problem the company is facing is that the price of tech equipment is soaring, and companies are trying to lock in purchases.
13:00 They are, at least for now, abandoning IBM's varied services, and the company reported results that were basically in line with its pre-announcement. So it reported earnings per share of 293, up 5% versus last year. But there was more bad news. The company cut its revenue forecast and is now expecting revenue growth to be up only in the range of 4 to 5%.
13:21 Part of the problem is that sales of data center mainframes were down 42% in the quarter, and infrastructure revenue was down 7% versus last year. IBM is clearly struggling. ServiceNow, one of the best software companies in the world, but the stock is down 38% this year because of the SaaS apocalypse.
13:41 The irony is that ServiceNow has executed really well, continues to grow, and shows no signs that AI is negatively impacting its businesses. But, fighting the AI narrative is like fighting a ghost. This quarter, once again, ServiceNow performed well. Earnings per share of 90 cents was up 10% versus last year, not bad.
14:03 Revenue reached 4 billion, and that was up 24% versus last year. I don't see any weaknesses in these numbers. Even so, the stock was down almost 4% on Thursday. This is fairly typical for how the market reacts to ServiceNow results. ServiceNow has had great quarters, and then seen the stock get swamped by the AI narrative. So, here we go again.
14:25 This coming Monday, by the way, on our free episode, we will explore who will be the winners and losers in the software world with tech analysts Dan Ives and Gil Luria. On Thursday, Blackstone reported. I'd say the results were good, but also mixed. Blackstone reported earnings per share of $1.52 versus $1.
14:43 21 last year, and versus $1.36 expected. Blackstone beat the street on higher transaction revenues and strong performance fees thanks to better-than-expected realizations. Fundraising was very strong at 68 billion. However, not everything is clicking right now, like slow base management fee growth and only modestly positive performance in certain asset classes, like real estate and credit.
15:09 Yes, the numbers were pretty good, but there are two major issues facing Blackstone and the private equity sector. The first, the time it takes to sell companies and give investors their money back keeps lengthening. And private credit's problems with software will really start to matter until next year when the refinancing cycle begins.
15:27 Nothing in today's Blackstone report alleviates any of these concerns. By the way, on the conference call, management bragged that it is the biggest financier of AI data centers. Given the intensity of the current AI debate, that may or may not prove to be a great bet. For me, the takeaway from this week's earnings results is that the terms of debate on AI have truly shifted.
15:53 A year ago, it was all rah-rah for AI. When companies raised their CapEx budgets, the market cheered. As the AI story has somewhat matured, the story has shifted. It's not all positive. The business has become capital intensive. Investors question whether there are any moats. China AI players have produced great models that are much cheaper and might create a price war.
16:17 Everyone is nervous, and that nervousness was on full display this week. That's why when Google raised its AI CapEx from 190 billion to 205 billion and posted negative free cash flow, the market did not cheer. Google was down 7% on Thursday, and because of the negative cash flow, Tesla was down 14 and 1/2% on Thursday as well.
16:37 The news from Google and Tesla caused Nasdaq to be down more than 2% on Thursday. Finally, Intel reported Thursday night, and the results were great. Earnings per share of 42 cents versus a loss of 10 cents last year blew away numbers. Revenue growth was the best it had been in 15 years. Sales in the data center segment soared 59% versus last year.
17:00 The stock was up after hours, but Intel's results, I don't think are going to quell AI nervousness. Given CapEx budgets, it would be surprising if Intel did not have a good quarter. But it's the AI CapEx budgets themselves that are making investors nervous. And now for the mailbag. Our first mailbag is from Pedro, who asks, "Quote, hi Steve.
17:22 Two questions for you regarding banks. One, what do you think of Bank of New York? I know it's a bit on a league of its own, but I would love to get your thoughts on how it compares to the other major banks. Two, if in general you see the banks as much safer now, have you considered adding one of them to your portfolio? What was the last time you were long a bank stock? That context would be very appreciated if you wouldn't mind sharing.
17:44 Thanks in advance." With respect to Bank of New York, I have to confess that I have never paid that much attention. Bank of New York and State Street are trust banks and operate in a world all their own. I have never found that much value to focusing on them. Like every other large bank, Bank of New York has done well.
18:04 It's up over 30% this year, but you could get the same performance by buying just about any other of the large cap banks. As for my portfolio, it's true that I don't own any banks right now. The last time I owned a bank was when I owned Citigroup, but that was before Jane Fraser became CEO, and I sold it after Citi experienced its umpteenth trading scandal.
18:22 I had the right thesis, but the wrong management. I should have bought an alternative, but I've spent most of my research time on tech stocks, which has proven to be a pretty good decision. My hesitation in buying bank stocks now is twofold. First, after experiencing a great run, they are at peak valuations. Perhaps more importantly, the strength of the investment banking cycle right now is heavily dependent on AI financing needs.
18:48 So, in a sense, owning Morgan Stanley, Goldman, Bank of America, Citigroup, etc. is just one more aspect of the AI trade. Simply put, buying banks to me does not provide diversification from tech. I am releasing a masterclass on how to analyze banks and how to value them in the near future, where I do a deep dive on the entire sector.
19:10 And our second mailbag is from Colin. Colin watched our recent premium episode with payments analyst Ken Sahousky. And in that episode, Ken mentioned that many hedge fund managers have shorted payments stocks as funding shorts. Colin asks, quote, I don't grasp the concept of shorting a stock for funding purposes while breaking even on a stock.
19:29 Great question. Let me answer it carefully. First, let's just review what happens mechanically when you short a stock. First, you borrow someone's stock. You then take that stock and sell it. Let's say you were shorting one share and the stock is $100. When you sell the stock, you now have $100 in cash.
19:48 Let's say the stock goes to 50. You then buy the stock back at $50 and give the stock back to the owner. You have made $50 less the fee you pay for borrowing the stock. That's how most people think of shorting. But let's say I think the stock I'm going to short is not going to do that much, but I want to use it as a way to fund buying something I really like.
20:11 So again, I borrow the stock and I sell it at $100. I now have $100 in cash. I take that $100 and buy stock in a company I really like. Essentially, I'm using the short to fund a long in a different stock. I'm hoping to make a lot of money on the long and either break even or make a little on the short. That's a funding short. This last Monday, July 20th, we released an interview with Ben Kallo, the sustainable energy and mobility analyst at Baird.
20:38 We discussed how the buildout of AI data centers has upended the entire sustainable energy landscape, creating a hypergrowth story. So check it out. And this coming Monday, July 27th, we will release an interview with Dan Ives and Gil Luria, two tech analysts who cover the full gamut of tech. We discussed how the debate around AI has shifted from being all positive to a much more nuanced discussion.
21:02 We talked about capital intensity, the lack of moats, the potential for an AI price war, and how real the threat is to software companies from AI. And we also discussed private equity's overexposure to software. So, please tune in. The best way to support the Real Vision Playbook is to subscribe to Substack and to YouTube.
21:24 Subscriptions are free, and we appreciate your support. And that's the wrap. This podcast is for informational purposes only and does not constitute investment advice. The hosts and guests may hold positions in stocks discussed. Opinions expressed are their own and not recommendations. Please do your own due diligence and consult a licensed financial advisor before making any investment decisions.