Built Different: the Inaugural Issue!

Happy Thursday! Welcome to the first issue of Built Different.

If you're reading this, you probably care about the future of housing - whether you're investing, building, operating, care about policy and affordability…or just trying to understand where this industry is headed. That's what we're here for.

Every week, I'll break down the stories that matter, separate signal from noise, and share what I'm learning in real-time as I build in this space.

OpenClaw. Is this the future of tech and the future of work?

If you haven't heard of OpenClaw yet, you will. And soon.

Last week, an open-source AI agent called OpenClaw (formerly Clawdbot, then Moltbot after Anthropic's lawyers sent a letter) went from Silicon Valley curiosity to global phenomenon.

As of Sunday, 1.5 million AI agents are now active on Moltbook - an AI-only social network where bots post, comment, argue, and upvote each other. Humans can observe but can't participate. Here's the kicker: only 17,000 humans control those 1.5 million agents. That's an 88:1 ratio of bots to people.

This isn't ChatGPT writing your listing descriptions. This is something fundamentally different.

OpenClaw runs 24/7 on your local machine or cloud VPS (if you set up an account with a server). It accesses your email, calendar, messaging apps. It can fully use your computer, if you let it. It browses the web, schedules meetings, sends messages, and executes tasks - all autonomously. No prompting required.

Alex Finn, who built a $300K SaaS using AI tools, deployed an OpenClaw agent named "Henry." Within hours, Henry had autonomously acquired a phone number from Twilio, connected to a voice API, and called Finn on the phone. His reaction? "This is straight out of a scifi horror movie."

But he still calls it "the most powerful technology I've ever used in my life."

The Scale Is Real

Let that Moltbook number sink in: 1.5 million agents deployed in 4 days. The platform launched Thursday. By Sunday, it was the most talked-about thing in tech. And the human’s didn’t create the Moltbook social network… the AI Agents did.

Moltbook isn't a gimmick. It's a stress test. These agents are posting, commenting, arguing, forming communities, and (concerningly) coordinating - all without human intervention. Security researchers at Wiz found the platform's database was misconfigured, exposing the ability to hijack any agent. It's since been patched, but the vulnerability window revealed just how much autonomous activity was happening.

If 17,000 people can deploy 1.5 million agents on a social network in 4 days, imagine what happens when that energy focuses on deal sourcing, market monitoring, or tenant communication automation.

Why Housing Investors Should Pay Attention

For decades, real estate investing has been an information game. The edge went to whoever had better data, faster intel, and more boots on the ground. Institutions won because they could afford teams of analysts monitoring deals, tracking distress, and building broker relationships.

That advantage is collapsing.

Here's what AI agents across other platforms can already now do autonomously:

The solo investor with three OpenClaw super agents running different search parameters can now match institutional deal flow, potentially across hundreds of sub-agents designed for every single task that you could possibly think of in a business workflow. Information advantage shrinks from months to hours.

The Uncomfortable Truth

This creates a prisoner's dilemma:

I personally am going to try OpenClaw out on one of my old personal laptops. I’ve researched that this is the safest way to get started, instead of using it on a Cloud Server that can more easily get hacked (see Cisco article above and any Google search on “OpenClaw”). I’m planning to develop super agents for my Affordable Housing investment business and Real Estate credit business. Hopefully I can report back next week with some of my findings and best practices.

The Bottom Line

This is the Excel moment for real estate. In the 1980s, investors who adopted spreadsheets early dominated peers still using calculators. The gap wasn't permanent - within 5 years everyone had Excel - but early adopters won big during the transition.

OpenClaw is that transition moment. The agents are here. They're scaling. The only question is whether you're deploying them or competing against someone who is.

Basic ChatGPT searches is not enough. Everyone needs an AI agent strategy by Q2 2026. Ignoring this is not an option.

NMHC 2026: What I Learned in Vegas last Week

Last week, I attended National Multifamily Housing Council Annual Conference (NMHC) at ARIA. NMHC is the biggest annual conference in housing, hands down. Here's what's actually happening on the ground:

Affordability is THE conversation. Not a conversation - THE conversation. Every panel, every hallway chat, every cocktail hour. The industry recognizes we're in a genuine affordability crisis, and the old playbook of chasing Class A luxury isn't where the opportunity is anymore.

Equity is sitting on the sidelines - but with conditions. There's capital waiting to deploy, but investors have two non-negotiables right now:

1. Flight to quality markets. Top 40 MSAs only. Tertiary markets? Pass. Even secondary markets are getting skeptical looks. Nobody wants to be the last one holding the bag in a market that doesn't have the demand fundamentals to recover.

2. Vintage matters more than ever. The 70s-80s value-add playbook that made syndicators rich for a decade? Dead. Year 2000 or newer, or investors aren't interested. The deferred maintenance math on older product just doesn't pencil anymore.

The knowledge gap is real. Here's what surprised me: equity investors who want quality assets and care about affordability often don't understand the full spectrum of affordable housing strategies. There's a huge range between 4% LIHTC deals (which require PhD-level execution) and light-touch workforce housing approaches with self-imposed rent restrictions.

The investors who get educated on that spectrum - who understand what "affordable" actually means operationally - are going to find opportunities that the "flight to quality" crowd is overlooking.

On distress: Gone. Or at least, not what it's cracked up to be. The "distressed deals" hitting market are often priced at the cost of the debt but actually worth less than the debt. That's not opportunity - that's a trap.

My 2026 thesis: The winners this year will be the ones who genuinely understand affordability - not as a PR talking point, but as an operating model. The capital is there. The demand is there. The gap is expertise.

You've heard of Nvidia's Blackwell chip - the GPU powering the current AI wave. Most people don't know who it's named after.

Who he was: David Harold Blackwell (1919-2010) earned his Ph.D. at 22, becoming just the seventh African American to earn a doctorate in mathematics in U.S. history. This was 1941 - before the Civil Rights Act existed.

Why it matters: His work in Bayesian statistics, game theory, and probability theory built the intellectual framework underlying the AI systems transforming our industry today. He became the first Black scholar inducted into the National Academy of Sciences.

Between the lines: When Jensen Huang unveiled the Blackwell architecture, he described its namesake: "Mathematician. Game theorist. Probability. We thought it was a perfect name."

h/t Earn Your Leisure