Closing Bell 11/5/26

Anthropic

Across the shows: 104 speakers in other episodes, 130 bullish and 87 bearish calls, past 30 days

• Dan Ives · Yorkville Ives & Co.Bearish

  Anthropic would be a negative call for Ives if public; valuation-dependent, given its capacity build-out needs.

  Ives says that if OpenAI and Anthropic were public today, he would be more negative on them than on hyperscalers. Those labs have to build out the capacity themselves, so they are more exposed to the slowdown than hyperscalers, which already have compute. He hedges that his view depends on valuation.

  “If OpenAI and Anthropic today were public companies, those would be the ones that I'd be coming here being like, relative... It depends on valuation. You have to deem me more negative on those names because they're the ones that have to build out the capacity.”at 11:03

• Brad Gerstner · Altimeter CapitalBullish

  Anthropic IPO may price at the low end, but demand remains strong and Altimeter will be a major buyer.

  Gerstner, an early Anthropic investor through Altimeter, says the market turmoil will make overall sentiment less euphoric and push the IPO toward the lower end of its range. He still sees a serious level of demand for the offering, and his enthusiasm for the company has not changed. He says Altimeter hopes to be one of the largest buyers.

  “So it will probably... You know, this pushes it a little bit more to the lower end of the range wherever it was going to price. But I'll tell you, there is a pretty serious demand for the offering. I know our enthusiasm for it hasn't changed.”at 12:23

• Alex Heath · Big TechnologyUncertain

  Anthropic will not delay its IPO; growth numbers look best now, though valuation may fall.

  Heath says Anthropic has to go now because its S-1 numbers, driven by heavy token spending and enterprise adoption, will look their best in this window. A delay would signal weaker conditions. Its valuation was built on future expectations, and those have shifted because it may have less compute for training. So the price could logically come down as it goes out.

  “They have to come out now because of the fact that their numbers are gonna basically look the best they ever will... I don't see them, actually pushing off.”at 15:33

NVIDIA

Across the shows: 73 speakers in other episodes, 68 bullish and 28 bearish calls, past 30 days

• Dan Ives · Yorkville Ives & Co.Bullish

  Nvidia's numbers are underestimated; demand still far exceeds supply, so the stock looks cheap for its runway.

  Ives cites demand-to-supply of about 13 to 1 seen on his Asia trip. Even if demand falls to 7 or 8 to 1 after data center cuts, demand still runs well ahead of supply. He believes Nvidia's street numbers are too low by 15-20%. He says that given where the stock trades, he cannot be more bullish on the runway through 2028-2029. He acknowledges Nvidia will have issues if the data center build falls short.

  “Still when I look at [Nvidia] numbers, I think they're underestimated by 15 to 20%. So when I look at the stock trading where it is, my view is that numbers are way underestimated.”at 9:03

Advanced Micro Devices

Across the shows: 12 speakers in other episodes, 12 bullish and 1 bearish calls, past 30 days

• Dan Ives · Yorkville Ives & Co.Bullish

  AMD is a buy on the chip side because the stock already prices in a haircut to AI demand.

  Ives says that compared with what these stocks are trading at, even after a 30% haircut to the backlog, investors should buy hyperscalers, big tech and some software names, including AMD on the chip side. The argument rests on the market already pricing in a cut to the AI build-out.

  “specifically some of the software names in [Nvidia], including AMD on the chip side.”at 10:27

Nebius Group

Across the shows: 5 speakers in other episodes, 2 bullish and 5 bearish calls, past 30 days

• Dan Ives · Yorkville Ives & Co.Bearish

  Nebius is on the wrong side of the AI buildout; neoclouds face massive consolidation.

  Ives says neocloud providers are clearly on the wrong side of the data center slowdown. He singles out Nebius as exposed. He expects a game of thrones with massive consolidation and musical chairs, with only a few neoclouds emerging as winners.

  “that's ones where they're clearly on the wrong side of this. To some extent, I'd say [Nebius]... massive consolidation musical chairs.”at 10:38

GE Vernova

Across the shows: 6 speakers in other episodes, 7 bullish and 0 bearish calls, past 30 days

• Dan Ives · Yorkville Ives & Co.Bearish

  Power players like GE Vernova face weaker demand because fewer data centers will be built.

  Ives says the data center slowdown means fewer data centers get built, which hurts the power equipment and generation players, naming GE Vernova first. The sentence is cut off in the transcript, so the negative view is implied rather than fully stated.

  “And then the power players from GE Vernuva to everyone else because you're gonna have less data centers.”at 10:50

OpenAI

Across the shows: 47 speakers in other episodes, 29 bullish and 39 bearish calls, past 30 days

• Dan Ives · Yorkville Ives & Co.Bearish

  OpenAI would be a negative call for Ives if public; valuation-dependent, given its capacity build-out needs.

  Ives says that if OpenAI and Anthropic were public companies, he would be more negative on them than on hyperscalers. OpenAI, like Anthropic, has to build out its own capacity, which leaves it more exposed if data center builds slow. He hedges that his view depends on valuation.

  “If OpenAI and Anthropic today were public companies, those would be the ones that I'd be coming here being like, relative... It depends on valuation. You have to deem me more negative on those names because they're the ones that have to build out the capacity.”at 11:03

Micron Technology

Across the shows: 23 speakers in other episodes, 25 bullish and 11 bearish calls, past 30 days

• Adam Parker · Trivariate ResearchBullish

  Micron is attractive on a three-month view; it can generate about $300B free cash over 78 quarters.

  With the market down and stocks treated equally, Parker looks for what is on sale and likes Micron, which he values at about $750B. He thinks it can generate about $300B of free cash flow over the next 78 quarters. He prefers buying businesses with immediate earnings power over newer cloud and rollout stories with less track record, so he is comfortable buying in on a three-month view and would dip a toe in, while balancing with health care, energy and Costco.

  “I look at Micron as one that I like, as you know, and it's what $750,000,000,000 market cap. I still think they're pretty good to do 300,000,000,000 in free cash in the next seventy eight quarters”at 44:51

AI infrastructure & data centers

• Dan Ives · Yorkville Ives & Co.

  Big tech capex growth peaks in 2027, earlier than expected, as 10-15% of planned data centers slip.

  Ives now expects 10-15% of planned US data centers to slip or not happen, possibly 15-20%. About 40-45 GW was due online and roughly 10-15 GW is now at risk or slow-rolled. Because capacity is constrained, he now sees big tech capex peaking in 2027, versus his earlier view of 2028 or 2029. He argues this means demand is still outpacing supply, so the market's overbuild and free-cash-flow worries are overdone.

  “It could be 15, 20% data centers that ultimately don't happen. But our view now is that CapEx for big tech, it peaks out in 2027 next year and from a growth perspective. Much earlier than you expected.”at 3:59

• Dan Ives · Yorkville Ives & Co.

  Hyperscalers already holding compute capacity are best positioned, so they are his preferred exposure despite the slowdown.

  Ives argues the market has worried about overbuilds, free cash flow and capital raises, and that this worry now gets thrown out. Hyperscalers already have capacity and more levers to pull, such as cutting spending, which could bring free cash flow back positive sooner. Capacity holders become the winners, while players that need to build out capacity are more threatened.

  “To some extent, that now really gets thrown out the window. The hyperscalers are in the best position in terms of how we change our view.”at 4:41

• Alex Heath · Big Technology

  Local voters restricting data centers, including in Ohio, will slow the buildout and shift competitive balance.

  Heath argues that the AI backlash, which has grown through polls, surprise data center projects and job fears, came due in the election. Voters in 38 jurisdictions, including more than a dozen in Ohio, voted to restrict data centers. Companies will eventually build, but sites already planned may now be off the table, so they have to restart planning. That slows the story and shakes up the competitive balance.

  “American people voted 38 jurisdictions, voted to restrict data centers, including more than a dozen in Ohio, which is a very big AI state.”at 14:08

• Tom Lee · Fundstrat Global Advisors

  AI selloff is overdone; demand is intact and a US slowdown is a speed bump, not a pothole.

  Lee argues AI demand will not slow because companies are still discovering and benefiting from it. There is no new AI competition that could wipe out the industry. A US slowdown would create opportunities for states that embrace AI and for countries with cheaper power. He thinks the repricing of these stocks is overdone, unless the speed bump becomes a pothole.

  “the demand side for AI isn't gonna slow because there's still a lot of discovery and companies are benefiting. And the supply side, there's not really new AI competition... this is sort of a speed bump question, but does that warrant the massive massive repricing of these stocks? I think that's overly done.”at 32:14

• Tom Lee · Fundstrat Global Advisors

  Companies will find ways around the blue-wave policy shock, such as retrofitting empty plants for data centers.

  Lee says companies were ingenious at handling the tariff shock, and earnings held up. He expects the same with the blue-wave policy shock. He has heard stories of empty industrial plants and office space being retrofitted into data centers, so power and gigawatts can still be built. He sees it as an opportunity for American businesses.

  “I think companies are gonna find ways to deal with this policy shock of the blue wave. And I think we even heard stories that, for instance, you may be retrofitting empty industrial plants and office space to put in data centers.”at 34:48

• Dan Greenhaus · Solus Alternative Asset Management

  Permitted data center builds will proceed; the real risk is 2028, and neoclouds shouldn't fall 15-20%.

  Greenhaus says the worry is that Democratic wins in Ohio, Texas and Georgia could halt data center construction. He argues the debate is not about 2026 or 2027, since already-permitted data centers will almost surely continue without a hitch. The real question is about 2028 for capacity beyond current permits. He does not think CoreWeave, Nebius-style neoclouds should fall 15-20% on this risk, and he doesn't expect Texas construction to stop. He also notes that after the election, data center concern would ramp up, which he had argued earlier, and that the market now has clarity on the topic.

  “already permitted data centers are probably almost surely going to go on without a hitch. The question is about twenty twenty eight.”at 43:06

Market sentiment, positioning & volatility

• Stephanie Guild · Robinhood

  The selloff is a needed reset, worsened by unwinding of record margin debt built up in the system.

  Guild says high-beta stocks had returns about three times those of low-volatility stocks, versus a normal one and a half times. Margin debt was at about 125% of its level, compared with roughly 75% at the last couple of peaks. The unwinding of that leverage is what is driving the selling. She thinks the sell-off was probably needed, but the margin buildup will make it more severe than fundamentals justify.

  “more margin is unwinding... margin was a [125%]. Normally, last couple of peaks, it was 75%. So I just think this is a sell off that we probably needed, but it's gonna be exacerbated by the margin that we've had building up in the system.”at 16:32

• Oliver Renick · CNBC

  After the VIX jumps five-plus points, S&P 500 returns over the next three and six months beat average.

  Using Bespoke data, Renick finds that when the VIX jumps at least five points, the median S&P 500 return over the next three and six months is above four and four-and-a-half percent, which he calculates is better than average. His caveat is that over a one-year horizon the outcome looks like any other day, and that over one month the average return is negative because of a few severe bear markets. He frames the current volatility spike as a more constructive setup for investors who zoom out over several months.

  “When the VIX jumps at least five points, the median S and P 500 return the next three and six months is more than four and four and a half percent respectively, which by my math is better than the average.”at 42:03

AI bubble & tech valuations

• Brian Levitt · Invesco

  This selloff is not a bubble bursting; valuation starting points were reasonable, so opportunities are emerging.

  Levitt argues AI demand over time is not changing, and the concern is the scale of the investment build-out. Rising rates over the year did much of the damage, not just the election. The equal-weight S&P 500 closed at an all-time high the day before, so the selloff is not broad-based. He calls it a reset given the large planned investment, not the end of a mania or the end of the business cycle.

  “This is not the bursting of a bubble. This is not the end of a mania. Valuations starting points were reasonable. Opportunities start to get created.”at 17:41

• Tom Lee · Fundstrat Global Advisors

  The AI trade has been repriced on unrealized fears; memory and chip supply remains very tight.

  Lee says the supply market for memory and semiconductors is still very tight, and multiples have already come in. The market has priced in skepticism about earnings and a belief that supply will loosen. If data center capacity is not built, compute will get more expensive, and the overall economics improve. So he says the AI trade has been repriced on things that have not happened yet.

  “There's still a very tight supply market for memory and semiconductors. That's still gonna be very tight, and the multiples have come in. So we've now priced in skepticism of earnings and then a belief that the market is actually gonna get loose.”at 36:08

US stock market outlook

• Stephanie Guild · Robinhood

  Value may outperform growth for a while, and tech may become value as cash flow matters again.

  Guild argues the market forgot about cash flow during the first six months of the year, when growth outperformed value for a long stretch. She thinks this selloff is a reminder to care about cash flow again. As a result, value may outperform growth for a while, and some tech may be re-rated as value.

  “It can't be all one trade... you need to care about cash flow as well... I actually think value may outperform growth for a while, and it still can mean tech, but maybe more tech becomes value.”at 21:04

• Dan Greenhaus · Solus Alternative Asset Management

  Stock prices often lead fundamentals, so today's selloff doesn't yet signal broad earnings cuts.

  Responding to the idea that there is no real reason for alarm until broad earnings are cut, Greenhaus says that prices often lead fundamentals rather than following them. He cites the saying that the stock market has predicted nine of the last five recessions. The implication is that the sell-off does not by itself show a deterioration in earnings, and that markets can be wrong about sentiment in either direction.

  “Quite often, prices lead fundamentals.”at 45:50

Software & SaaS

• Dan Ives · Yorkville Ives & Co.

  Software will outperform semiconductors as AI use cases keep building out.

  Ives believes the use cases for AI will keep building out. He cites Palantir and Snowflake, and says enterprise demand is about four to one, with software customers telling him to wait until 2027 or early 2028 for projects. He argues software is the better place than semiconductors for the AI theme.

  “I think, ultimately, it's software over semis because my view that the use cases are gonna continue to build out in terms of the capacity that's there.”at 4:41

AI: overall outlook

• Liz Thomas · SoFi

  Elections won't stop the AI trade; AI is still early in its innovation cycle.

  Thomas says the idea that elections will stop the AI trade is not what will happen. She argues AI is still early in its innovation cycle. She also thinks the market is overshooting to the downside, and that the semiconductor trade may be directionally correct in the short run.

  “But the idea that elections are gonna stop the AI trade, I don't think is is what's going to happen. I do think that we're still early in the innovation cycle of AI.”at 18:51

Commodities (broad)

• Liz Thomas · SoFi

  In a stagflationary late cycle, commodities and energy should do well.

  Thomas says the market may be in classic late-cycle behavior, not recession. With the Fed in a hiking posture and inflation hot and rising, she thinks the environment could be stagflationary. In that setting she expects commodities and energy to do well, and yield-curve shifts to follow as the market watches whether the curve flattens or steepens.

  “We're in a period where we're... I don't think we're at the end of a business cycle. I don't think we're on recession watch by any means, but we might be in classic late cycle behavior. And what I would expect in that environment, especially if it's a stagflationary late cycle environment, is that things like commodities, energy do well.”at 19:21

US growth & recession risk

• Brian Levitt · Invesco

  Market turmoil reflects a slowdown within an ongoing cycle, not the end of the business cycle.

  Levitt grants that big rate moves bring cracks, but investment-grade spreads have not blown out, and inflation expectations are still contained. The Fed is not raising rates enough to invert the yield curve. Leadership has rotated to staples, real estate, utilities, health care and Treasuries, which he reads as expectations of a slowdown rather than a recession.

  “In terms of spreads blowing out significantly in the investment grade space, you haven't seen that. Inflation expectations are still generally contained. So I don't think the Fed's raising rates to where they're inverting the yield curve. So all of that says to me cycle continues.”at 20:24

IPOs, M&A & deal activity

• Leslie Picker · CNBC

  Anthropic IPO fees per bank are modest; follow-on IPOs and debt deals matter more to big banks.

  Picker says fees on big IPOs are usually capped around $100 million per bank. That is a lot, but it won't make or break Anthropic's revenue. What matters for the big banks is the IPOs that follow, other debt issuance tied to the deal, and wealth-management revenue from IPO windfalls. Investment banking fees were already tracking below expectations.

  “fees capped maybe around a $100,000,000 per bank. So that's a lot, but it's not necessarily going to kind of, you know, make or break this company... So it's the IPO. It's other IPOs that follow.”at 27:45

US rates & the Fed

• Tom Lee · Fundstrat Global Advisors

  Productivity data makes the September hike look a mistake; a December hike probably won't happen.

  Lee says today's productivity number changes the narrative on inflation risk and on Fed hikes. He thinks the September hike looks like a mistake and a December hike probably doesn't happen. If the AI buildout slowdown becomes an employment risk, the Fed put would come into play, which he sees as an inherent stabilizer.

  “Today's productivity number means I think the September hike looks like a mistake. The December probably doesn't happen, but then I think the Fed put comes into play if it becomes a [an] employment risk story.”at 33:44