The market is paying for AI stories, but not without wobbles
Palantir Technologies closed at $186.03 on Friday, October 16, a drop of $14.47 (more than 7%) that left the shares lower for a fifth straight session. Only days earlier they had hit a 52-week high of $212.94.1 The fall came on a broader technology sell-off, so it was not specific to Palantir.1 The more interesting point is what analysts did on the same day. At least two of them stayed with the stock or became more positive, and that contrast with the traditional defense sector is what this piece examines.
One caution first. The source material for this piece is strong on analyst opinion and announcements and weak on hard Palantir financials. It contains no Palantir revenue, profit or contract figures, so we make no claims about them. The only audited numbers it holds belong to Goldman Sachs, and we use them only for context, as described below.
What the analysts said
Ryan Canady upgraded Palantir to Buy. He concluded that AI applications built by customers themselves can remain complementary to Palantir's platforms rather than displace its core product.1 That is a direct answer to a common worry: that big customers will build their own AI tools and stop paying for Palantir's.
Bernard Zambonin remains constructive. He points to customer expansion and net dollar retention (a measure of whether existing customers spend more over time) as central to the bull case. He also acknowledges that the elevated forward valuation makes continued execution essential.1 In plain terms, the shares are priced for things to go well, so there is little room for disappointment.
The topic brief behind this piece also mentions a Goldman Sachs upgrade. The supporting documents do not describe it in enough detail for us to report its terms, so we do not.
A week of new partnerships
The announcements that fed this optimism came in quick succession in mid-October:
- On October 13, Ondas and Palantir announced Regna AI, a standalone private company meant to help governments, businesses and institutions build and run AI models in sovereign environments (ones kept under a country's or organisation's own control). Palantir is expected to provide model-training infrastructure and engineering.2
- On October 14, Customers Bank announced a collaboration with Palantir to develop an AI-oriented banking operating model. Its own forward-looking statements flag implementation risks, reliability concerns and contractual flexibility.3
- Also on October 14, Surf Air Mobility signed its fifth commercial customer for OperatorOS, software powered by Palantir. Surf Air receives a percentage of revenue from trips managed through the platform.4
These are announcements, not results. The Customers Bank statement is a useful reminder: the bank itself says implementation could be harder than hoped.3 None of the documents give financial terms for any of the three deals, so we cannot say how much revenue they might add.
The other side: defense stocks under a cloud
J.P. Morgan describes a challenging third-quarter earnings season for aerospace and defense companies. It cites slowing airline traffic growth, uncertainty over U.S. military spending and lingering supply-chain constraints as weights on investor sentiment.5 In an October 8 report, its lead analyst for the sector, Seth Seifman, said most companies should deliver solid results and express confidence in underlying demand. He added that good earnings may not be enough to reverse the sector's recent stock-market weakness.5
In the same report, J.P. Morgan downgraded Leidos. Seifman's view was that Leidos' relatively low valuation does not justify an Overweight rating, because other aerospace and defense stocks also offer substantial potential returns.5 That is a relative judgment, a choice of better places to put money, rather than a forecast that Leidos will fail. The same report expects Howmet's 2026 earnings estimates to rise, with management confident about growth in aerospace and industrial gas turbines.5 So the weakness is not uniform across the sector.
Politics adds uncertainty. The November 3 congressional midterm elections could change control of Congress and constrain military spending growth, though some uncertainty may fade once the vote is over.6 Palantir serves both government and commercial clients,1 so it is not insulated from this theme. The documents do not say how its revenue splits between the two, so we cannot say how exposed it is.
Where speculation shows up
The appetite for AI-linked stories also shows in more speculative names, and the evidence there is mixed.
D-Wave Quantum shares fell about 10% on July 28 to $17.64, a day after a 20% surge, and were down 25% over the month.7 Analyst Melissa Tucker pointed to a 171-times price-to-sales multiple. Projected 2026 revenue is near $35 million, and her fair-value estimate of $9.50 per share implies roughly 40% downside. She cited a delayed technology roadmap and reliance on a supplier that IonQ is in the process of acquiring.7 Marc Gerstein, by contrast, sees D-Wave's combination of two quantum-computing methods as a strategic advantage over single-method rivals.7 That is a genuine disagreement between analysts, and we do not resolve it.
On Bloom Energy, analyst Nicholas Amicucci argues that its ability to supply reliable, dispatchable power to a volatile demand profile sets it apart from competitors.8 A separate Motley Fool piece reports that NuScale Power (SMR) shares are down nearly 40% in 2026, while a Bank of America analyst, Rinny Singh, reiterated a buy rating with a $12 target.9
Context from checked numbers: Goldman Sachs
The one set of audited figures in our evidence concerns Goldman Sachs, a bank whose analysts feature in this cluster of calls. These figures do not measure Palantir's prospects. They show only that an institution issuing calls is itself doing well. Earnings per share were $22.87 in fiscal 2023, $40.54 in 2024 and $51.32 in 2025.10 The latest quarters continue the climb: Q1 2026 earnings per share were $17.55, against $14.12 a year earlier,10 and Q2 2026 were $20.98, against $10.91 in Q2 2025.10 Cash fell over the same stretch, from $241.6 billion at the end of fiscal 2023 to $164.3 billion at the end of 2025.10 That is a change in the balance sheet, and the documents do not say why.
How much to trust the sources
Most of the analyst reporting here comes from the NewsEOD feed. In our measured checks, 57% of 881 claims from that feed held up against their source of truth.1,5,7 The Bloom Energy item scored lower, at 45% of 520 checked claims.8 The Motley Fool items scored 22%, though on only 18 checked claims, which is too small a sample to judge by.9 We therefore treat every analyst call as one person's opinion, reported secondhand. Readers should do the same.
What to watch
- November 3: the midterm elections, and any signal on military spending growth and the coming budget talks.6
- Whether the Palantir pullback continues. The shares fell for five sessions from a high of $212.94, and Zambonin himself notes that the valuation leaves little room for slips in execution.1
- Follow-through on the partnerships. Look for disclosed terms from Regna AI, Customers Bank and Surf Air, and watch the implementation risks Customers Bank flagged.2,3,4
- Defense earnings. Seifman expects solid reports but warns they may not lift the stocks.5
The evidence supports a narrow conclusion. Analysts are rewarding AI adoption stories at the same time as they are choosier about government-linked defense names. It does not show that Palantir is a bargain or a bubble. That would take revenue and contract data this dossier does not contain.
