Jace Kohlmeier

The sole legitimate purpose of an economy is to improve lives. It is not the purpose of a life to improve the economy.

@rushicrypto I was also thinking about this today I mean, it’s not like ten years are gone exactly, but it sure has been one heckuva annoying background soundtrack to an otherwise pretty excellent movie

We only get around 75 years to live… if we’re lucky. And Trump has taken 10 of those years from our peace. That’s not small. That’s a huge part of our life gone.

Blue state playbook: create a supply problem, throw subsidies at the symptoms, then raise taxes to pay for it all. Because you increased demand instead of supply everything gets more expensive. See: housing, energy, healthcare, childcare in CA vs TX. Eventually, people leave (or don't come), including many wealthy folks in the tax base (the top 1% pay 45% of taxes in CA)… but you keep spending, so taxes have to keep going up. Washington, New York, California are all moving to increase taxes, meanwhile, red states are moving in the opposite direction, competing for the same wealthy residents that blue states are pushing away, and they are attracting them, and many of them are bringing jobs with them… and people can live there because it’s cheap. I spent some time in Texas this year and it’s easy to see why their model works- it’s the real abundance state… They just let people build and don’t get in their way. Often ugly, but it is cheap to live there, and that makes a huge difference. They have TONS of immigrants for this reason. They have jobs and it's cheap to live there- pretty simple. Despite all the subsidies for immigrants in California, and lack of them in Texas, immigrants choose Texas over California... Cheap housing does more for quality of life than any government program. Same pattern with solar. California subsidizes demand while making it painfully slow to permit and install, while Texas just makes it easy to build. Texas generates more solar power than California and is adding new capacity at 7x the rate. Why don't I move? SF has network effects in tech and is the most beautiful state in the US. I just wish it were better governed!

Pretty sure this is the wildest thing I've read in 2026... and it's been a wild year! It's incredibly well-written and has interactive explainers - recommend at least a skim It's about LLMs embedding computers in the model. Here were some bangers 🧵

meanwhile corticallabs.com/cloud

I’m deeply uncomfortable with this. War is noisy, chaotic, messy, bloody, smelly and terrifying - after the event for me, rather than at the time. When you’ve had someone’s head splashed over you, smelled the stench of someone’s flesh burning, seen corpses ripped apart and scattered across the ground, a dead persons guts spilled across the road and covered with dirt and flies…then you may realise that it’s not like a computer game, Hollywood film, or about cheering on a football team. War is on another level. At the effects end it is serious beyond the comprehension of those who’ve not experienced it intimately. It scars you. When it comes to it, someone has to stand between those who would do harm and those to whom they would do it. But things that glorify war or portray it as a sport or game, as this video does, concern me.

As I have been saying repeatedly, this principle is entirely defensible, and this is the single best articulation of it anyone in the administration has made. The way to enforce this principle is to publicly and proudly decline to do business with firms that don’t agree to those terms. Cancel Anthropic’s contract, and make it publicly clear why you did so. Right now, though, USG’s policy response is to attempt to destroy Anthropic’s business, and this is a dire mistake for both practical and principled reasons.

No administration of my lifetime has been more antagonistic toward the private sector, free markets, and free trade. All the warnings of socialists and Democratic overreaches and government intrusion — it's all worse in reality right now than ever before. Just remarkable.

@beffjezos "Let's be real" --> makes outlandish counterfactual claim that can never be verified or tested 👌

Let's be real if the Dems had won OpenAI and Anthropic would have already been nationalized into a surveillance panopticon for them to stay in power forever

Some good points here. They would be better points if our government and military leadership was more morally sound.

Okay, so you’ve got Anthropic explicitly refusing to agree to terms that would strip AI safety guardrails and fully allow use of Claude in mass surveillance and autonomous weapons, WITHOUT human oversight. Something I’d say is pretty ethical of them not giving the military unrestricted use of high-tech AI without limits. And then you have a president calling that left-wing and unpatriotic. Because of course, it’s very patriotic to remove all the guardrails in mass surveillance….. You’re a real country hero when you totally abandon your ethics on a highly sensitive matter it seems 😂 What a joke.

It’s extremely good that Anthropic has not backed down, and it’s siginficant that OpenAI has taken a similar stance. In the future, there will be much more challenging situations of this nature, and it will be critical for the relevant leaders to rise up to the occasion, for fierce competitors to put their differences aside. Good to see that happen today.

@xEBITDA They also made more than one subtle allusion to make the product (not the line!) work reliably, and dodged on field data.

Given the credibility problem and the lack of specifics given to investors to underwrite a target gross margin at scale, I have concerns that all the revenue growth could just be a ramp to nowhere. I want new, credible management and more to go on than "look, we're a structurally profitable business (conceptually)"

$EOSE earnings take: If you strip away every promise management has ever made, and just look at the numbers, the growth profile is actually impressive. Revenue: - 2024: ~$15M - 2025: ~$115M - 2026 guide: ~$300M That’s not incremental growth. That’s a company scaling from pilot phase to real commercial deployment. Margins are improving. Volume is scaling. The product is clearly finding market demand. So why is the stock down and sentiment so bad? Because the disappointment isn’t about the growth profile. It’s about credibility. Management consistently guided higher than what ultimately materialized. Important distinction: The issue isn’t that $EOSE isn’t growing. It’s that management consistently over-promised and under-delivered. Without those aggressive projections, we likely never would have seen $19. But we also wouldn’t have seen this kind of violent drawdown. Ironically, the hype cycle had one major benefit: It allowed the company to raise substantial capital and strengthen the balance sheet to survive the ramp. Now we’re left with something interesting: A company with a real growth trajectory… But a management team that needs to rebuild trust. Or better, be replaced. The next phase for $EOSE isn’t about vision. It’s about execution and credibility. If they simply guide conservatively and hit numbers consistently, sentiment can reset, and the multiple can follow. The story isn’t broken. The messaging was.

@daniel_koss @spacanpanman What is your source for the 20-30% claim? AFAIK They haven't explicitly guided to a margin range since 2020

@spacanpanman Management is targeting 20% to 30% gross margins at steady-state scale. Obviously now everything they say sounds like "trust me bro".

$EOSE as you can see, this was clearly an amazing quarter. The issue is NOT bad execution (but yes, there were some hiccups). The issue is crazy overpromising leading to totally unrealistic expectations. Now people are disappointed, even when the company does really well. That's why I LIKE the pretty low $300-400m guidance for 2026. I think they learned their lesson. From the outside this of course looks bearish, as expected guidance was much higher. But again, I like it. To regain investor confidence, they must now start to underpromise and OVERDELIVER. Not the other way around.

@bert_gilfoyle rev recognition timing?

$EOSE In no world can they ramp to 4GWh by EOY and make only $300-400m. Totally incongruous.

I really want people to see the story above the story here, which is that whether you're reading Citrini, or listening to Jamie Dimon at a cocktial party, the conversation about AI is a marketplace of competing science fiction narratives. That's not to say I think the technology is a parlor trick. But rather that the level of uncertainty is so high, and the quality and supply of real-world, real-time information about AI's macroeconomic effects so paltry, that very serious conversations about AI are often more literary than genuinely analytical. And I think that observation sets up another important point: I feel lucky to be able to have conversations about the frontier of AI with executives and builders at frontier labs; economists at AI conferences; investors in AI; and other AI folks at off-the-record dinners where important truths can theoretically be shared without risk. I can't emphasize enough that "nobody knows anything" is about as close to the reality here as three words are going to get you. Nobody what's going to happen this year, or next year, or the year after that. There is no secret cigar-filled room of people who have unique access to some authentic postcard from the future. When you drill down underneath the bluster, the boosterism, the fear, the anxiety, what's there at the bottom is genuine uncertainty, a vacuum into which storytelling is flooding. The frontier labs don't really know what they're building exactly, and economists don't really know how to model the thing they claim they're building (genuine recursively self-improving AI agency isn't really analogous to something we know about). I wish more people talked about and thought about this subject thru that sort of lens: We're trying to model the economy-wide effects of a technology whose properties the frontier labs can't even really describe yet. Whatever you think about AI today, be prepared to change your mind soon.

Didn't read. Don't need to. My confirmation bias is LIT UP.

Fun. Only took a minute to setup and try. I asked a panel of expert swing traders to give me advice. The summary was fine, but the full raw dialogue gave lots of interesting detail and specific suggestions.

@XCapitalMgmt Really solid commentary. Thanks for the level-headed balance.

$IREN earnings: *How's Demand? Strongest to-date. All the hypers raised capex guidance beyond expectations. *How's Price Action? Brutal. Market wanted deal, market didn't get deal, price got cracked. Bought more after hours last night. *How Are The Fundamentals? Better than ever, more below. Specific fundamentals: *Canada - PG air cooled work being done shows they executed what was in their control, now just waiting on GPUs. $0.4B contracted so they should hit $0.5B Q1 guide. No mention of PG liquid cooled expansion which was annoying but there was a lot to cover. And commentary about air cooled demand specifically picking up suggests they should meet or beat the $3.4B year-end ARR guide. *H1-4: On schedule, great. $3.6B GPU DDTL from Goldman and JPM at <6% rate is excellent and hopefully a template for future GPU financings, which it sounded like. And the fact that they're going to look at financing for the long EUL datacenter buildings will further improve the economics of the $MSFT deal and put more cash back on the balance sheet. *Next-up deals (e.g., H5-10 and Sweetwater): This was what the market wanted but didn't get. A few select quotes then I'll add a few notes: "We are in multiple advanced negotiations, and at this point, demand is not the constraint for us. The focus is on choosing the right long-term partnerships that support durable platform-level growth." "We're advancing multiple active negotiations with a range of hyperscale and non-hyperscale counterparties, and what we consistently see is that customers are focused on partners who have secured power, can deliver full data center infrastructure on a defined timeline, and who can grow with them and scale over the long term. In other words, time to data center has become the key decision point in many of these commercial discussions. That dynamic plays directly to our strengths." "We're also seeing hyperscalers and leading enterprises actively pursue both liquid and air-cooled GPU deployments as they work to accelerate rollouts. The increased focus on air-cooled deployments aligns extremely well with our existing footprint of 810MW of already operational air-cooled data centers." [Unclear if this means Mackenzie/Canal or if H5-10 could have an air cooled component.] "One of the contracts we are negotiating at the moment is a multi-billion-dollar contract where we need to bring a software solution. So it is not holding us back. It will not hold us back." "We've had an ongoing dialogue on [Sweetwater] for 12, 18, 24 months with various parties. And as we've tried to reiterate, it, it needs to be the right deal." "We're really pleased around the GPU financing result, because it's kind of ticked that box. We're now on to the next one, and it also helps catalyze a lot of these other customer negotiations we're having and advance into the next phase, because we need capital, and you can't build without capital. So the GPU financing is now done. On to the next one. We've got the capacity, we've got the customers, and the demand and the negotiations underway. And as Anthony said, we've got what we see as really good access to capital at the moment." "Having now absorbed the capital requirements associated with our Microsoft deployment, we're able to focus on converting a broader set of advanced customer negotiations into contracted revenue." New ERCOT batch process has spooked the market but they were clear the power's secured and only impact would be load ramp timeline. They'll have 1.4 GW energized in Q2 (annoying they didn't say April as they have before) but they can't actually use the full 1.4 GW until they sign a deal and build the DCs so the batching process should be a non-issue. Bears will FUD mgmt. and not believe them, but I'm not playing that game - we'll know in a few months. Possible a hyper made it a last minute closing condition to confirm batch zero in which case we could get a deal shortly after that's made clear. We'll see. As for the deals themselves, they need to deliver H5-10 and/or Sweetwater 1 by May earnings or the market will puke given the commentary above. If no deal by May earnings, the question will be: why hasn't the right deal surfaced and will it ever? I don't think they could get anything done for H5-10 or Sweetwater without finalizing the $MSFT GPU financing. You have to de-risk delivery for your biggest client first. And now that's done, they've said they can convert advanced negotiations into contracts. The market will expect it and punish price if they don't. *Next-up sites: New 1.6 GW in Oklahoma is fantastic but market will value at $0 until we see Sweetwater 1. Market has no idea how to value 1.4 GW at Sweetwater 1 so no chance OK gets valued by the market until Sweetwater deal is announced. They also seemed to suggest there were more ERCOT sites beyond SW 1 & 2 that could be in batch zero - very interesting but same conclusion as above. *Cap stack: Great progress given GPU financing and good that they're looking to finance the H1-4 DCs to further improve levered economics. Most interesting commentary was that they alluded to having similar conversations with $NVDA as $CRWV for different financing structures (they didn't say $NVDA directly but there aren't many alternatives). Given the emphasis on maintaining appropriate debt/equity mix and given the fact that they're pursuing DC financing which would return cash to the balance sheet but increase debt, the bread crumbs above make me think some type of $NVDA or hyperscaler equity investment in $IREN is still very possible. Could be part of the "select corporate level initiatives" or the financing discussions tied to next-up deals in the quotes above. Exciting but let's see. Overall solid earnings. Noisy on the historicals but that was expected w/ the balance sheet changes and mining to AI Cloud transition. If we anchor back to analyst expectations, Goldman assumed $3.8B revenue by FY28 in their December $39 / share initiation and $IREN just firmed up guidance for $3.4B ARR by year-end (which will translate to revenue in 2027). That $3.4B doesn't include Sweetwater, H5-10, or Oklahoma. Need deals to raise targets materially. TLDR market wanted deal and didn't get one, so we're down bad. I expect that "deals or down" setup will be even more intense for May earnings. Time to monetize H5-10 and/or Sweetwater. I believe they will - I'll wait.

@SCurveCapital 💯

$IREN put the price action aside which is out of our control. Can someone prioritize making the website not brick for next EC? Not a good look for a CSP, especially.