Capital on the deals beat is being repackaged around proof rather than potential. In three stories logged on this beat recently, the money - whether venture debt alternatives, Navy co-investment, or a hobbyist's bot - flows to ventures that can show a working result first. The unifying thread is that the terms of engagement are being rewritten by whoever can demonstrate traction, and that has direct consequences for how US technology companies structure deals and how US consumers ultimately feel the effects.
Repayment Tied to Outcomes
Skalar's newly launched model, reported by Crunchbase News, gives startups capital to fund sales and marketing and lets them repay it out of the revenue generated by the customers acquired with that capital. That is a structural change in the venture debt conversation. Traditional venture debt is a loan against the balance sheet, repaid on a schedule regardless of whether the growth it funded materialised. Skalar's model, as described, ties repayment to the specific revenue stream the money was meant to create. The lender is no longer underwriting the company's general prospects; it is underwriting the unit economics of a customer acquisition motion.
For US startups, that reframes the diligence burden. A founder pitching this kind of facility has to show that a dollar spent on sales and marketing returns more than a dollar of revenue, and has to accept that repayment is measured against that cohort rather than against a quarterly cash position. The model is described as fairly straightforward though somewhat unusual, and that is the right read: it is simple in concept and demanding in execution. Companies with clean attribution and reliable payback windows will find it attractive. Companies whose growth is harder to trace to a specific spend will find it awkward, which is precisely the sorting mechanism the structure implies.
The Navy Wants Skin in the Game
TechCrunch reported that Navy CTO Justin Fanelli is pitching investors on co-investing alongside venture capitalists rather than funding early research himself, citing recent buys including a $562 million autonomous refueling deal and an updated wish list spanning AI to quantum. That is the same logic as Skalar's, applied at a different scale. The Navy is not proposing to carry the risk of early-stage research on its own balance sheet. It is proposing to stand beside private capital, which means private capital does the early screening and the Navy buys into what has already survived scrutiny.
The direction of travel matters for US technology companies. A co-investment posture signals that the Department of Defense wants deal flow that has already been de-risked by commercial investors, and it wants to move on defined programs rather than open-ended research. The $562 million autonomous refueling deal is the proof point: a concrete capability, a defined price, a completed transaction. The wish list - AI, quantum - tells founders where the Navy expects to be a buyer, but the co-investment framing tells them the Navy would rather be a later-stage participant than a first cheque.
For the US market, this concentrates early-stage risk in the private sector while giving the government an option on the winners. That is a meaningful change from a posture where public money seeds the earliest work. It also creates a two-step market: commercial investors set the terms and the valuation, and government procurement follows. Founders building in the listed areas should read the co-investment pitch as a signal about the capital stack they will need to assemble before a government buyer engages.
A Bot as a Proof of Concept
Tom's Hardware reported that a Reddit user built a GPT-6 Astra-powered bot to beat Balatro's Gold Stake Black Deck, leveraging Python for numerical tools and beating the hardest difficulty repeatedly. On its face this has nothing to do with acquisitions or deals. Read as a signal, it belongs on this beat. A single enthusiast, working on a consumer game, produced a system that reliably solves a defined, difficult problem using general-purpose AI plus purpose-built numerical tooling. That is the same pattern of demonstrable, repeatable results that the Skalar model and the Navy's co-investment posture both reward. The bot is a proof of concept, not a product, but it is the kind of proof that the current deal environment is set up to notice.
