Michael Horne

Advisor. Investor. Operator.
Ex-Arm/SoftBank. $1.5B+ in licensing & royalty deals closed.
Building and scaling revenue for deep-tech founders.

[Michael Horne @launchright]

Fair on the vested interest, I help startups jumpstart sales. But fewer seats to sell doesn't mean no sales. People still buy cars, food, healthcare; that demand doesn't go away because a company runs leaner. And something the doom case misses: agents are becoming buyers too. They sign up for tools, pay for data, spend real money. New market, no human interface needed.
Fair to say where humans earn will change, though. That's how every tech shift looks at first: jobs vanish in one column and show up in another. The work doesn't go away, it moves, and there's usually more of it once the dust settles. Painful for whoever's caught on the wrong side, no question. But "no humans" has never been how it plays out.
It's really a shift in who does what, and who's even doing the buying. More sales, not fewer.

[Michael Horne @launchright]

@tomkeene Disagree. A salary is a subscription too. The real question isn’t price, it’s output per buck spent. Same price for the same output? You’re doing it wrong. Same price for 10x, sometimes 100x? You’re buying outcomes no human can match. I’d take that all day long.

[Michael Horne @launchright]

Not buying it. A salary is a subscription too. So the real question isn’t price, it’s what you get back per buck.

Pay an agent a salary’s worth in tokens and get the same output? Sure, that’s a bad deal, and you’re probably doing something very wrong.

The deal is pay same price, 10x the output. Sometimes 100x. At that point cost savings isn’t the point. You’re buying outcomes a human can’t physically match. You’d take that all day long.

[Michael Horne @launchright]

@TimurNegru Many times, wonderful morning hike from Camogli and then delicious lunch at Puntetta or da Giovanni, hang on the beach after and boat back to Portofino or SML.

[Michael Horne @launchright]

And it's now coming for physical engineering. In fact, it's already here. What traditionally takes months of engineering work to do (design an electric motor, layout a complex chip) can now be done in days, even hours. Buckle up.

[Brett Caughran @FundamentEdge]

A big pivot from Ken Griffin on AI:

“Number one is, in the last few months, there has been a step change in the productivity of the AI toolkit. It is profoundly more powerful than it was just nine months ago.

And for us at Citadel, that has allowed us to unleash a much broader array of use cases for AI. And it has been really interesting to watch, to be blunt, work that we would usually do with people with masters and PhDs in finance over the course of weeks or months being done by AI agents over the course of hours or days.

These are not these are not mid-tier white collar jobs. These are like extraordinarily high skilled jobs being, I'm going to pick a word, automated by agentic AI. And I gotta tell you, I went home one Friday actually fairly depressed by this because you could just see how this was going to have such a dramatic impact on society.

When you witness it in your own four walls, when you see work that used to be man years of work being done in days or weeks, it's like, wow, like that's the first time I've seen real impact in our four walls.”

This echoes my own experience with agents and the conversations I am having with students, friends & clients. The toolkit has dramatically transformed and it feels like in finance, for the first time, AI is real.

[Michael Horne @launchright]

Not practical, but interesting. Maybe a better way to summarize this:

"I built an agent loop that found a low-effort bounty, wrote a working PR, and earned $16. The novelty is that I didn't write the code. The economics don't scale, but the capability is real and will matter when bounty supply goes up or agent cost goes down"

[Michael Horne @launchright]

@ajlamesa What is typical in the world is not inevitable. Japan culture has proven it.

[Michael Horne @launchright]

None of these go away on their own.

You can't out-wait them.

The founders who clear them treat commercial readiness as a system, not a guess.

If any of these hit close to home, follow us @launchright. We help deep-tech founders close real deals before the runway closes them.

[Michael Horne @launchright]

When the anxiety hits at 3am, every deep-tech founder is thinking the same 9 things.

None of them go in the board deck.

Here they are. And what actually works.

🧵

[Michael Horne @launchright]

9: "Enterprise procurement will eat us alive."

Legal reviews, security questionnaires, 12-month cycles. Nobody gave you the rulebook.

Build the contract pack early. Mutual NDA, MSA template, security responses. Procurement isn't the enemy. Being surprised and unprepared is.

[Michael Horne @launchright]

@BorisVagner Hermes is a great option, solid.

[Michael Horne @launchright]

@brivael Spend enough time around deeptech founders and your point becomes obvious. This is the work. The hardest problems don't solve themselves, and they don't get funded by accident. Someone has to back the people willing to try.