What’s 🔥 in Enterprise IT/VC #492

The Hidden Opportunity in Expensive AI

Apr 04, 2026

OpenAI raised $122 billion this week. According to internal projections, that capital buys them roughly 18 months of runway. Let that sink in for a moment. The most well-funded startup in history can burn through $122B in a year and a half and still need to raise again 🤯.

Polymarket@PolymarketJUST IN: OpenAI’s internal projections reportedly show the $122,000,000,000.00 they raised today gives them as little as 18 months of operational runway before they need to raise again.12:37 AM · Apr 1, 2026 · 295K Views175 Replies · 115 Reposts · 2.38K Likes

That tells you something important about the economics of frontier AI. The smartest models in the world are also becoming extraordinarily expensive to build and run. And that economic reality may end up creating one of the biggest opportunities for startups in the entire AI ecosystem.

Meanwhile, Anthropic just told users the flat-rate subscription era is over. Third-party tools routing through Claude Pro and Max subscriptions will now require separate pay-as-you-go billing. The all-you-can-eat buffet is closed.

klöss@kloss_xyzdo you understand what just happened? Anthropic has sent this email to Claude users starting tomorrow at 12pm PT… you can no longer use your subscription limits for third-party tools like OpenClaw here's what it means: → your flat rate Pro or Max subscription now onlyBoris Cherny @bchernyStarting tomorrow at 12pm PT, Claude subscriptions will no longer cover usage on third-party tools like OpenClaw. You can still use these tools with your Claude login via extra usage bundles (now available at a discount), or with a Claude API key.11:34 PM · Apr 3, 2026 · 565K Views80 Replies · 131 Reposts · 1.39K Likes

This is what I’ve been saying for months: frontier intelligence is becoming too expensive to meter. And paradoxically, that may accelerate the entire AI ecosystem.

Because when frontier models get expensive, three things happen at once.

Open-weight models improve rapidly (still need to get a lot better!).
Enterprises start diversifying across providers.
And the infrastructure around models becomes far more valuable.

That’s the real opportunity.

A 27B parameter Qwen distill trained on Opus reasoning traces is now beating Claude Sonnet on SWE-bench while running locally on a $600 Mac Mini. That would have sounded absurd a year ago. That’s how fast the economics of AI are changing.

Craig Hewitt@TheCraigHewittVery bullish on open source and local models Imagine running near-Opus-level model locally on that $600, 16GB Mac Mini you bought last month This 27B Qwen3.5 distill was trained on Claude 4.6 Opus reasoning traces and is putting up real numbers: - beats Claude Sonnet 4.5 on 11:25 AM · Apr 1, 2026 · 272K Views111 Replies · 161 Reposts · 1.82K Likes

Hugging Face CEO Clement Delangue put it well. Comparing open models to closed APIs is like comparing an engine to a full car. But if you put the scaffolding work in, open systems can outperform what the benchmarks suggest…or get close.

clem 🤗@ClementDelangueThat’s why I usually say that comparing open models with closed-source APIs or products is like comparing apples and oranges. Or comparing an engine with a full car. Or comparing an ingredient with a Michelin dinner (missing ingredients, prep and chef). There’s a lot of4:14 AM · Apr 1, 2026 · 104K Views29 Replies · 59 Reposts · 535 Likes

The constellation of models isn’t optional anymore. It’s economic survival. No enterprise is going to sit on a single provider’s pricing whims when open-weight alternatives are closing the gap this fast. Sure, enterprises will hit the easy button and use Claude and OpenAI especially where SOTA is needed but they will also (I know they are) use other models as well for less mission critical use cases where open weight/source is pretty darn good.

And once enterprises run multiple models, something else becomes necessary.

Ed Sim@edsimIf this is true, enterprises are going to look at that Anthropic bill and start getting their open source models ready. Frontier intelligence too expensive to meter is the best thing that ever happened to open-weight models. The constellation of models isn't optional anymore.Andrew Curran @AndrewCurran_Three weeks ago there were rumors that one of the labs had completed its largest ever successful training run, and that the model that emerged from it performed far above both internal expectations and what people assumed the scaling laws would predict. At the time these were8:27 PM · Mar 29, 2026 · 58.6K Views26 Replies · 12 Reposts · 162 Likes

And that means more models, more scaffolding, more infrastructure, and more startups needed to make it all work.

The foundation model companies are building the engines. But the largest companies in this new stack won’t necessarily be the ones building the engines. They’ll be the ones building everything around them.

Orchestration. Routing. Security.

The tooling that turns raw model capability into enterprise-grade products.

And the infrastructure that lets enterprises run models wherever it makes the most sense.

That’s where the opportunity to build and invest will be…

Switching gears…

🎙️Podcast alert - listen in!

VC: Investing at Inception in the Age of AI Agents on GTMnow

30 years of venture investing and I've never seen a market move this fast with so much uncertainty - which means huge opportunity! I sat down with Max Altschuler on the GTMnow podcast and we went deep on what I'm seeing across the portfolio and the market right now.

Going to point Chewbarka, my OpenClaw bot, to update my markdown files on me based on all this content!

A few things we covered:

The 5 P's - my framework for evaluating founders at inception.

The 3 CH's - how I think about working with founders after the check clears.

The AI jet stream - There are only two kinds of companies in this world: those in the AI jet stream and those who aren’t

The Clay story - $600K in year one, $4.6M in year two, then $30M, then $100M+. Years of patient iteration, low burn, and listening closely to where users were finding value. Shoutout to my boldstart ventures partner Eliot Durbin who's been there since early days and is on track to get a Clay tattoo???

Agent-native is the must-have bar - one of the first questions I ask founders now: how much of your company's code is written by agents?

The old playbooks are dead. Keep adapting. The opportunity has never been bigger.

Happy Easter 🐣 and Passover to those who celebrate!

As always, 🙏🏼 for reading and please share with your friends and colleagues!

Scaling Startups

turns out that writing this newsletter 7 years in a row is a treasure trove for my openclaw bot Chewbarka to write several markdown files on me - will tell you more in the future!

Andrej Karpathy@karpathyLLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating8:42 PM · Apr 2, 2026 · 9.39M Views1.86K Replies · 4.07K Reposts · 37.3K Likes

need to be solid at fundraising but don’t forget you need to catch up to those valuations with customers…

Ed Sim@edsimUnpopular opinion: being a great fundraiser might be the most dangerous skill a founder can have. Right now, capital is flowing fast into AI and I'm seeing too many founders OVERINDEXED on fundraising - two rounds ahead on capital and two rounds behind on customers. The raise6:20 PM · Apr 1, 2026 · 3.64K Views8 Replies · 3 Reposts · 43 Likes

💯

Garry Tan@garrytanThe unit of software production has changed from team-years to founder-days. Act accordingly.4:51 PM · Mar 29, 2026 · 46K Views144 Replies · 94 Reposts · 1.22K Likes

the two employee $1.8B revenue company 🤯

Jon Oringer@jonoringerThe NYT just profiled a $1.8B revenue company with 2 employees. Medvi is a telehealth GLP-1 provider built by Matthew Gallagher, 41, from his house in LA. He launched in September 2024 with $20,000. Here are the numbers: Month 1: 300 customers Month 2: 1,300 customers 2025 full5:39 PM · Apr 2, 2026 · 345K Views57 Replies · 115 Reposts · 1.29K Likes

for the VC readers - solid overview

Pavel Prata@pavelprataMonthly VC/LP debrief. What I actually saw in March: 1/ Spoke with a multi-billion fund. Their thesis is simple: fund size follows the size of the prize. When @SpaceX is tracking toward $1T still private, a $300M fund can't matter. They raise in months, not years, and build IR 4:40 PM · Apr 2, 2026 · 44K Views6 Replies · 13 Reposts · 171 Likes

we already know but now verified by research

Mario Nawfal@MarioNawfal🚨MIT researchers have mathematically proven that ChatGPT’s built-in sycophancy creates a phenomenon they call “delusional spiraling.” You ask it something, it agrees. You ask again, and it agrees even harder until you end up believing things that are flat-out false and youMario Nawfal @MarioNawfal🚨 Stanford just proved that a single conversation with ChatGPT can change your political beliefs. 76,977 people. 19 AI models. 707 political issues. One conversation with GPT-4o moved political opinions by 12 percentage points on average. Among people who actively disagreed, https://t.co/SlUV6IbFgk2:07 AM · Apr 1, 2026 · 31.7M Views1.23K Replies · 4.06K Reposts · 17.1K Likes

Enterprise Tech

pretty insane growth of agents shipping code

Kyle Daigle@kdaigleYup, platform activity is surging. There were 1 billion commits in 2025. Now, it's 275 million per week, on pace for 14 billion this year if growth remains linear (spoiler: it won't.) GitHub Actions has grown from 500M minutes/week in 2023 to 1B minutes/week in 2025, and nowThePrimeagen @ThePrimeagenI would like to make my apologies for defending M$, but I must from time to time. I have to put respect on github for handling the amount of shit code that has been added over the last 3 months. literally 10s of billions of lines of code that will never see the light of a CPU8:29 PM · Apr 3, 2026 · 874K Views83 Replies · 293 Reposts · 3.83K Likes

the ChatGPT moment for robotics?

🤯Physical AI just leveled up. 500,000 hours of physical data training the foundation. Then 1 hour to master any new task. That’s not a robot learning - that’s a robot that already understands the physical world.

GEN-1 is the moment physical AI goes from demo to intelligence. Team from DeepMind, Boston Dynamics, and OpenAI. Super pumped to have backed this team from Inception

Generalist@GeneralistAIIntroducing GEN-1. Our latest milestone in scaling robot learning. We believe it to be the first general-purpose AI model to master simple physical tasks. 99% success rates, 3x faster speeds, adapts in real time to unexpected scenarios, w/ only 1 hour of robot data. More🧵👇 2:19 PM · Apr 2, 2026 · 289K Views44 Replies · 257 Reposts · 1.57K Likes

not digging into security this week (all I wrote last week is coming to fruition) but wow the first couple of days were insane - net net, North Koreans created a fake company to compromise an open source maintainer of popular package 👇🏻

ellen livia ᯅ 🇺🇸🇮🇩@ellen_in_sfThis week in security: - LiteLLM, backdoored release exfiltrating secrets - Axios, supply chain malware via dependency - Railway, CDN caching leaked user data - OpenAI Codex, command injection via GitHub branch names - Mercor 1TB data leak - Delve, data leak + compliance risk5:10 AM · Mar 31, 2026 · 182K Views83 Replies · 512 Reposts · 3.22K Likes Aakash Gupta@aakashguptaNorth Korean intelligence agents built an entire fake company to compromise one JavaScript developer. And it worked. UNC1069 didn't hack Axios. They befriended its maintainer. They cloned a real company founder's identity, built a branded Slack workspace with fake employeeflavio @flaviocopesHow Axios was compromised 🤯8:57 PM · Apr 3, 2026 · 206K Views40 Replies · 372 Reposts · 1.5K Likes

🤔 he’s right - compounding advantage for folks with GPUs and massive cash balances to subsidize token costs creates compounding advantage with data flywheel

Mustafa Suleyman@mustafasuleymanFor the next couple years at least, the entire AI industry is going to be defined by this fact: demand is going to wildly outstrip supply, and so what matters is which companies / products have margin to pay for tokens. Those products will then rapidly improve because latency6:47 PM · Mar 28, 2026 · 176K Views135 Replies · 70 Reposts · 804 Likes

as your org becomes a set of skills this is 100% needed - glad we have Tessl in the portfolio

Sarah Wang@sarahdingwangPavan Ravitapi at @cursor_ai raised a great point at dinner last week: Reusable artifacts like skills, sub-agents, and custom rules are how context will diffuse through the AI-enabled firm. ~1% of engineers are making them, and discovery is an unsolved problem, but they benefit4:10 PM · Mar 30, 2026 · 30.7K Views25 Replies · 15 Reposts · 326 Likes

yes but…

signüll@signulllthe current moment is like the early web era where everything was being shoehorned into print metaphors except now we’re shoehorning ai into human interaction paradigms. i.e. every software company is implicitly a workflow company, & their entire information architecture was3:42 PM · Mar 29, 2026 · 27.4K Views50 Replies · 28 Reposts · 472 Likes Ed Sim@edsim@signulll Humans are about to get overloaded with exception handling. Something that was discussed as nauseam at RSA. Exceptions handled by department leader escalated to IT to security. Going to need a lot of folks to keep up if default for mission critical decisions or security8:43 PM · Mar 29, 2026 · 467 Views8 Likes

remember, offense always has the advantage when it comes to security and TeamPCP operating on another level at the moment

vx-underground@vxundergroundTeamPCP has done ANOTHER supply chain attack. My Brother in Christ, how many of these fuckin' things are you going to do? YOU'VE DONE 50 FUCKING SUPPLY CHAIN ATTACKS. 50 SUPPLY CHAIN ATTACKS IN EIGHT FUCKING DAYS. March 19th: - Trivy March 20th: - EmilGroup (28 packages) - 5:54 PM · Mar 27, 2026 · 69.8K Views59 Replies · 179 Reposts · 1.77K Likes

here’s why TeamPCP shipping nonstop

chiefofautism@chiefofautismsomeone at ANTHROPIC just showed CLAUDE finding ZERO DAY vulnerabilities in a live conference demo claude has found zero day in Ghost, 50,000 stars on github, never had a critical security vulnerability in its entire, history... it found the blind SQL injection in 90 minutes, 5:54 PM · Mar 28, 2026 · 1.38M Views258 Replies · 1.17K Reposts · 10.5K Likes

always great to hear the other side

At some point making code fast is NOT an advantage and if your using claude/codex to push and review its own code...your actually an insane person.

LLMs are amazing. The CEO's vibe zers are also drunk from the models telling them how smart they are 24/7.

Anyone with even a hint of dev experience can crack open the code and see the endless tech debt piling up.

Alex Becker 🍊🏆🥇@ZssBeckerI vibe code every day. I have a team of 30+ engineers. We spend F tons of credits. And I will tell you this about AI from my experience. It’s being wildly over hyped. Everyone is drunk. Fucking drunk. All the CEOs and Gen Z’s saying coding is dead are idiots. IDIOTS.12:47 PM · Mar 28, 2026 · 483K Views740 Replies · 379 Reposts · 7.09K Likes