Edition 1

Companies don't just evolve relative to their environment, but also against each other. That press release (written by AI, of course) that your business just published announcing their new AI agent? Your competitor will launch theirs next week. The fastest adoption of a commodity input wins nothing, is not differentiating, and shouldn't have you feeling ahead.

In addition to this pressure, AI actually makes your average day harder, not easier. AI makes your day denser, replacing mindless mechanical work with higher-stakes orchestration, strategic choices, and other activities with much higher cognitive requirements. This increased cognitive density is tiring and dangerous, consuming the resources you could use to run a race where you can actually win.

To avoid the Red Queen Trap, it's not about adopting the commodity the best or the fastest. It's about recognizing that commoditization has changed the game on the field, and focusing on building differentiated innovation further up or downstream. As commerce moved online, while everyone rushed to create the best website, Amazon asked themselves what happens when digital distribution goes to zero, which led them to pour billions of dollars into distribution infrastructure to build the greatest flywheel in the history of retail.

In conclusion, "we added an AI agent" is table stakes, not a moat. Running a race in the exact area that is being commoditized is how you end up running twice as fast to achieve nothing. The winners innovate around the commodity input, not on it, and build around what is possible, with an eye to what becomes scarce, when intelligence is a commodity.

Link → Running Faster to Go Nowhere: The AI Adoption Trap, BuccoCapital, Educated Guess

No one actually has figured out what the software engineering organization looks like in the AI-native era, once you let go of the legacy mindsets, the legacy cultures, and the routines that worked five years ago. Across one VC's portfolio, however, a trend is starting to emerge.

One "run-the-business" side optimizes for reliability and steady cash flows. The other "transform-your-business" side exists to take bets and open optionality. These are two completely different optimization parameters, and they haven't seen any being or team handle both at the same time successfully. It's like asking a VC to take bets, but at the same time optimize the portfolio for steady cash flows.

"When you ask an engineering manager to make the existing platform more reliable while also asking them to ship a wild new bet that could break it, you're asking them to be schizophrenic. They can't do it."

"We think it's actually one of the more hidden mechanisms behind Clayton Christensen's innovator's dilemma. The same person being asked to deliver both ends up choosing the safer outcome, almost every time, because that's how human incentives work in incumbent contexts."

"The takeaway we keep landing on is that AI native isn't just a tooling decision. It's an organizational design decision."

Link → The Engineering-Org of the Future: Bet-Taking Factories and Commercial AI Engineers, Patric Hellermann, LinkedIn

The thesis: in AI, everything visible gets easy to copy, so the harder thing to copy becomes the institution underneath that can turn work into a compounding system no other company can reproduce.

"The best companies have always known that people are not an input to the company, but rather are the company." Great companies are organizational inventions, creating a new kind of institution around a new kind of work, and in doing so, making a new kind of person possible.

OpenAI, Palantir, Tesla, and more are examples of this (read the long-form article for more detail).

Link → The shape of the company is becoming the moat, Jaya Gupta (via @dshan), X

5x to 10x ARR is no longer the default. Elite public SaaS trades at 3x to 8x revenue, with ServiceNow near 4.8x and HubSpot near 2.7x.

Buyer bucket sets the range: value buyers 1x to 2x, PE-quality 2x to 6x, growth equity 4x to 8x, strategics 7x to 12x.

What lifts a business above "solid": an owned workflow, proprietary data, measurable AI leverage, and real meat left on the bone (price increases, margin, cross-sell) for a buyer to underwrite.

Link → Luke Sophinos' Post, Luke Sophinos, LinkedIn

Software investing used to be as simple as looking at ARR growth and retention, and spitting a valuation out, with public and private markets moving in lockstep. That world is gone.

AI capabilities are uncertain, and there is a shift from clear and well-understood business models to new paradigms where long-term business models are far from certain.

The best way for investors to navigate this uncertainty is to embrace it. This is what finance as a discipline is all about, taking calculated risks against an uncertain future. Conviction and a willingness to be wrong beat false precision. It is not the case that uncertainty is now temporary. Yesterday's certainty was the real mirage.

The days of Benjamin Graham, when you got paid to own companies at a discount to asset value, are dead. That was the value era. The days when you paid to own high growth companies with a strong Rule of 40 are dead too. That was the growth era. The AGI paradigm is an attempt to recreate the sense of certainty we used to have as investors, one that has slipped through our fingers. Instead, we are entering the era of uncertainty, where there is no option but to take real hard risk, and be right.

Link → After Certainty, David Cahn

Our companies, like the many family businesses in America, are policy takers not policy makers. They must operate and react in an environment where planning is hard, agility is an imperative, and leadership is key. In periods like this, businesses that adapt faster than their environment, stay grounded in what does not change, and invest for the long-term win. That is the lens through which we view both our companies and Kanbrick itself.

"We think a key question is not whether AI is real, but how quickly it diffuses from Silicon Valley to Main Street... Diffusion of AI at scale is not simply a better model problem... Technological progress is not the same as deployment."

The electric-motor lesson: winners redesign the whole system rather than bolt AI onto the old process. "When electricity first arrived in factories in the 1890s, most owners did the obvious thing: they ripped out the steam engine and dropped in an electric motor. It took a generation before manufacturers realized the real opportunity was redesigning the factory."

Bezos and Buffett often focus on what won't change, and Charlie Munger's principle of inverting is particularly helpful here. At the first principles level, a few examples: customers will still prefer better, cheaper, and faster; incentives will still drive behavior; and moats and capital allocation will drive enduring success. Most importantly, leadership still matters, and matters most in dynamic environments.

"Evolution and capitalism have many parallels, and we are big believers that the best enduring companies evolve faster than their outside environments. This becomes especially important in a time of disruptive change where AI is accelerating decision loops."

Link → 2025 Annual Letter, Kanbrick