🔹 Thursday’s selling had nothing to do with bonds. It came from a private company’s revenue line.

Stocks set fresh records as Hegseth confirms the Iran ceasefire and oil retreats nearly 4%.

🔹 Thursday’s selling had nothing to do with bonds. It came from a private company’s revenue line.
The Bull Investor
Daily Brief · Friday, October 9, 2026
  ◆  

Rates Got Cheaper. Stocks Fell Anyway.

Thursday's selling had nothing to do with bonds. It came from a private company's revenue line.

Thursday's Session

The Close

Thursday split the market in two. The Nasdaq Composite fell 345.35 points, or 1.25%, to 27,193.34. The S&P 500 gave up 36.41 points, or 0.47%, to 7,765.36. The Dow, meanwhile, finished higher — up 51.77 points, or 0.10%, to 51,231.64. One index was down more than twelve times as much as another was up.

What separated them was artificial intelligence. The Financial Times reported that OpenAI's annualized revenue is running near $50 billion — roughly $20 billion below figures widely cited in September — and everything attached to the AI buildout sold off. Oil climbed too, with Brent settling 3.7% higher.

Presented by our partner Sponsored*

Doroni text logo

Man walking near futuristic flying vehicle

You were promised a flying car.

For decades, it lived in cartoons and sci-fi movies.

This summer, that changed. The FAA’s new MOSAIC rule took full effect, and for the first time, a two-seat personal aircraft has a clear path to certification.

One company has been building for that moment for nine years.

Doroni Aerospace started in a South Florida garage with a 3D printer. Today its prototype is cleared by the FAA for flight testing and has completed 70+ successful test flights. Its chief engineer previously held leadership roles at Eviation, the company behind the first all-electric airplane.

Meet the H1-X: a two-seat, all-electric aircraft designed to take off vertically, fly at 120 mph, recharge in 25 minutes and be flown after about 25 hours of training.

Watch the H1-X in action »

And people are already lining up:

  • 600+ pre-orders, representing $240 million+ in potential revenue
  • $45 million+ raised from 16,500+ investors
  • Its last raise filled up in under 100 days

Morgan Stanley projects urban air mobility could become a $1 trillion market by 2040, with $457 billion tied to carrying people, the segment Doroni is built for.

Invest at $3.30 a share
Minimum investment: $1,003.20. The more you invest, the more bonus shares you receive:
$2,500 or more 5% bonus shares
$5,000 or more 7% bonus shares
$10,000 or more 10% bonus shares
$15,000 or more 15% bonus shares
$20,000 or more 18% bonus shares
$25,000 or more 20% bonus shares
Already an investor or H1-X pre-order holder? You get an extra 10%.

October 22 is the last day to invest at $3.30. On October 23, the share price changes.

Invest at $3.30 Before October 22
One Number, One Sector

One Big Thing

OpenAI is a private company. You cannot buy its shares, it files no quarterly report, and nothing it did on Thursday showed up in any index. Yet a single revised figure about its sales took more than a percent out of the Nasdaq. Following that chain is worth the effort, because it explains how a great deal of today's market is actually wired.

OpenAI buys an enormous amount of computing power — chips, servers, data-center capacity, electricity. The firms it buys from are public. So when the FT put its annualized revenue near $50 billion rather than the roughly $70 billion that had been doing the rounds, traders did not reprice OpenAI. They repriced its suppliers. Revenue is what eventually has to pay for all that hardware, so a smaller revenue figure quietly lowers the ceiling on how much hardware anyone can reasonably expect it to keep buying.

OpenAI's spending is not a line in OpenAI's accounts. It is the revenue line of companies you may already own.

The chipmakers took it hardest, with Intel, Micron, AMD and Broadcom all falling somewhere between roughly 3% and 5% during the session. Nvidia closed down 2.94% at $230.48. Among the big platforms, Amazon fell 2.3% and Microsoft 1.4%, while Apple, which sells no data-center hardware to the AI labs, rose 1.1%. That last contrast is the whole story in miniature: the further a company sits from the data center, the less Thursday touched it.

The Risk That Got Cheaper

The Chart That Explains It

Fall in Treasury yields, Oct 7 → Oct 8, 2026 · basis points
2 year
 −2
5 year
 −4
10 year
 −6
30 year
 −7
Source: US Department of the Treasury, daily par yield curve.

Wednesday's villain became Thursday's relief. Every maturity cheapened, and the long end — the part that had just hit a 24-year high — cheapened most. Borrowing costs fell across the board, and stocks still dropped. Whatever was wrong on Thursday, it was not the bond market.

What Nobody Can Audit

The Deeper Read

Before anyone concludes the AI business is shrinking, look at what actually changed. Reporting on the gap suggests a good part of it is a measuring difference, not a collapse in sales. The higher figure counted the full value of certain partner transactions; the lower one counts only OpenAI's own share. Same company, same week, two defensible numbers roughly $20 billion apart.

That is the uncomfortable part, and it has little to do with OpenAI. A meaningful slice of the earnings growth investors are paying for in public chip and cloud companies rests on purchase plans made by private firms that publish no audited accounts. There is no 10-K to check, no auditor's signature, no standard definition of the revenue everyone is extrapolating from.

Contrast that with the same day's bond market, where the Treasury sold 30-year debt and the demand was measurable to the decimal: bids covered the offering 2.54 times against a 2.41 average for the last six sales, and indirect buyers took 72.3% versus a 69.1% average. Investors could see exactly who showed up. On the AI side, they are reading press reports about a number nobody is required to define the same way twice.

A word from our sponsor Sponsored

I've just revealed a powerful investing strategy that boils down to " Sell This, Buy That."

It's a way to rid yourself of overpriced AI stocks before the tech trade breaks down this summer...

And instead move that money into smaller, lesser-known names that are showing real potential to dethrone the "Mag 7".

I even give away a Hotlist and Hitlist of buy and sell ideas that you can act on right now.

Like my recommendation I call "an upgrade to Tesla stock." It's a little-known company that just inked a groundbreaking partnership with the king of AI, Nvidia. This deal virtually hands this under-the-radar firm the keys to the self-driving industries' biggest customers, putting them miles ahead of Tesla in the autonomous vehicle race.

That's why I want to put this stock on your radar before markets open.

You can get the name and ticker symbol here.

For Your Portfolio

What This Means For You

If you hold a total-market or S&P 500 index fund, you own the AI buildout whether or not you ever formed a view on it. Thursday is a reasonable way to find out how much. A day when the Dow rises and the Nasdaq drops 1.25% is a free measurement of your own exposure — look at what your accounts actually did, not at the headlines.

The rest of the economy looked steady. New claims for unemployment benefits came in at 197,000 for the week ending October 3, close to the 198,000 four-week average, which is not what a weakening labor market looks like. Borrowing is still the pinch point: Freddie Mac's Thursday survey put the 30-year fixed mortgage at 7.40%, up from 7.28% a week earlier and 6.30% a year ago.

Next week brings something firmer than a press report. September inflation data arrives Wednesday, October 14, and the Fed meets October 27–28.

Stepping Back

The Long Game

Every market cycle has one number everybody leans on and nobody can verify. In 1999 it was eyeballs. In 2006 it was the assumption that house prices do not fall nationally. Today it is how much a handful of private AI companies will spend, and for how long. I have no idea whether that number holds. What I do know is that when a thesis rests on figures outside the audited record, the sensible response is to size the position so you can be wrong without it mattering much — not to guess harder.

  ◆  

— Anthony

* This is a paid advertisement for Doroni Regulation A offering. Please read the offering circular at https://invest.doroni.io/