Open in interactive viewer → charts, metric popovers & call review

Blue Owl Technology Finance: A 14% Yield, a GPU Bet, and the One Corner of AI That's Falling Apart

OTF's largest new investment lands in AI compute just as neocloud credit reprices violently — while the dividend-coverage date quietly slips another year
OTF · Earnings Call · 2026-11-05

The pitch versus the tape

Craig Packer closed Wednesday's call with a number pitched to stop you in your tracks: a “dividend yield north of 14%” — Craig Packer, Chief Executive Officer · 2026-11-05. Set that against the tape. OTF is down 42.5% from its June 2025 peak of $16.56, down another 10.6% over the last ninety trading days, and sits at roughly 57 cents on the dollar against the $16.62 net asset value the company just reported. Every BDC in America trades at a discount; this one trades at a chasm. The bull case is that the assets are fine and the market is wrong. Packer's evidence: public software equities “have rallied approximately 35% from their April lows, reversing nearly all the decline earlier this year” — Craig Packer, Chief Executive Officer · 2026-11-05. Non-accruals are 1.1% of fair value across three names, over 80% of the book is senior secured at a 40% weighted-average LTV, and PIK is 11.5% of investment income with 98% of it structured at origination rather than bolted on through amendments. That is the recurring credit quality argument, and the numbers back it up on the surface. But the shape of the earnings underneath is less flattering. Net income peaked at $180M in Q4 2025 and has been declining since; the June quarter came in at $154M. Net investment income printed $0.29 per share against $0.30 last quarter. The base dividend is $0.35. That is a six-cent hole funded by spillover income of roughly $0.35 per share, propped up by realizations such as SpaceX and $225 million of buybacks over the past year.

What's genuinely new: GPUs on the balance sheet

The vocabulary shifted this quarter in a way that matters. Last quarter's top keywords were defensive — credit quality, public loans, private credit, AI disruption. This quarter the top of the list is offensive: opportunity set at number one, then higher spreads, target leverage, and deal flow. Management stopped defending the book and started selling the setup. The concrete new thing is GPU financing — a theme with no prior history in this company's keyword record. Erik Bissonnette: “The $2.4 billion transaction included $1.1 billion provided by Blue Owl, including $180 million from OTF” — Erik Bissonnette, President · 2026-11-05. The borrower is IREN, the facility is senior secured on the GPUs, amortizes fully over a short tenor, and is contractually structured for a low-teens return. Digital infrastructure is now nearly 2% of the portfolio, with a stated ambition of mid-to-high single digits. Here is the tension, and it is not small. OTF is pushing into the exact corner of the AI complex that is currently in freefall in public markets. Over the last 30 days the global tape shows AI data centers down 30.6% with 11 advancers against 55 decliners; data center capacity is down 43% over 90 days. IREN itself sits among the heaviest decliners in the global factor set. Bissonnette acknowledged the environment in real time:

Obviously, there has been some pretty dramatic adjustments over the past few days in terms of unsecureds in the neocloud environment, as well as their equities. What we know is that the need for capital is enormous, and we will continue to be involved in financing these businesses.

Erik Bissonnette, President · 2026-11-05
The defense is structural rather than fundamental — Packer: "low double digits and very short duration… we think it is extremely defensive." That may be right. It is also the argument every lender makes right before a collateral class gets repriced. OTF's own new keyword record suggests the market has not yet decided which way this cuts.

The promise that keeps moving

Covering the dividend has been the central guidance anchor for four straight calls, and the date has drifted. In February, Jonathan Lamm said the company was “very, very much on track in terms of delivering the NIIs for the dividend… by the end of this year” — Jonathan Lamm, Chief Financial Officer · 2026-02-24. By May, the framing became "we have not changed our view of the ultimate outcome — just the timing." By August, Erik Bissonnette was saying “we expect to cover that dividend by the middle of next year” — Erik Bissonnette, President · 2026-08-06. This week: the same phrase, mid-2027, unchanged. Credit where due — it has stopped slipping. But the arithmetic has not gotten easier. Getting there assumes roughly $1.2 billion of net deployment and leverage rising from 0.94x toward 1.15x. That requires roughly double the recent run-rate of gross originations in a quarter where commitments were $441 million against $335 million of repayments. Management's answer is that net leverage is room to run, not a symptom — and that spreads on new paper now average 5.8%, 120 basis points wider than the 4.6% seen in Q4 of last year. There is one piece of genuine numeric support: liabilities-to-assets sits at 49.9%, up 11 points year over year but still well below the historical 2022-23 peak above 49% on this basis, and common equity of $7.5B still stands behind it. The fund is not stretched. It is just not earning.

What quietly disappeared

The most telling absence is the old language. AI disruption — a top-ten keyword two quarters ago — is gone from the current set. So is the explicit life sciences theme, even though the quarter's headline realizations (Catalent, XOMA Royalty) came precisely from there and Bissonnette flagged life sciences as one of three deployment avenues. When a company stops naming a pillar while still citing its cash flows, the pillar has been subordinated to something newer. That something is compute. The honest read: OTF is a levered, floating-rate lender to software businesses whose stock market believes the loans are worth less than management says. Management has responded not by shrinking but by adding duration-short, structurally-protected exposure to the one theme in the market currently repricing downward hardest — and by telling you the spread compensation for doing so is the best it has been in years. Both things can be true at once. That is what makes the next two quarters a real test rather than a re-run of the last two.