release edition [069]

read time [9 minutes]

Welcome to The Multifamily Download, a weekly newsletter where I provide institutional insights to help you build an exceptional Multifamily career.

Forwarded this email? Subscribe here.

Today at a Glance:

• AI Goes Institutional: Anthropic-Blackstone JV
• Late Adopters: Why CRE Is Catching Up Fast
• Career Capital: The Skill That Matters Most
• Weekly Listen: Jon Gray, Blackstone

TMD 069 is brought to you by:

On our last acquisition, the lease audit alone took two analysts the better part of a week. That was 365 units. Multiply that across a larger portfolio and diligence becomes the bottleneck to closing. Rely is fixing this. Upload your lease files, and Rely's AI handles classification, extraction, and audit flags. Most audits complete in under 30 minutes. The result? Your team reviews clean outputs instead of building spreadsheets from scratch. 100% document coverage. Source-linked verification. Accuracy you can trust. Try your first audit free →

​

Today's edition kicks off The AI Operator, a new series that I plan to build upon as time goes on.

The conversation around AI in the Multifamily sector is too important, and moving too fast, to confine to a single newsletter or moment in time.

Let me know what you think of today's newsletter, and if there are any specific AI topics that you'd like to see in future editions of The Multifamily Download. Enjoy!

The Anthropic-Blackstone JV And What It Means For Multifamily

The biggest story of last week didn't come from Yardi or RealPage. It came from a press release.

On Monday, Anthropic, the company behind Claude, announced a new enterprise services firm alongside Blackstone, Hellman & Friedman, and Goldman Sachs, with additional capital from Apollo, General Atlantic, Leonard Green, GIC, and Sequoia.

Total committed capital is roughly $1.5 billion, with Anthropic, Blackstone, and Hellman & Friedman each putting in about $300 million.

The new firm will embed Anthropic's engineers directly inside the company and deploy Claude across the consortium's portfolio companies first.

Mid-market companies are next.

Real estate is one of the named target sectors.

In TMD 052, my seventh prediction for 2026 was that AI would go mainstream in CRE. I laid out the 5-thought adoption framework in TMD 057, the 5 governance questions every operator should ask before deploying AI tools in TMD 059, and confirmed the prediction as on-track in my Q1 scorecard in TMD 062.

From where I sit, this announcement is the most concrete structural confirmation of my TMD 052 prediction yet.

There are three things worth noticing here:

1/ Capital is going to the implementation layer, not the technology.

Blackstone, Apollo, and GIC are some of the most thoughtful capital allocators in the world. They are not buying Anthropic, but rather they are buying the company that puts Anthropic to work inside operating businesses.

This alone tells us where they think the bottleneck is and where they think the returns are greatest.

The model itself is a commodity over time, but the forward-deployed engineering capability that makes the model usable inside a real business is not.

Why? Because the first mover advantage is asymmetric when technology is accelerating this quickly.

For context, Anthropic has shipped 100+ Claude features thus far in 2026.

The possibility that a company can reach escape velocity to become "uncatchable" by the competition in this rapidly evolving post-AI world is more real than ever before.

2/ Blackstone owns a lot of multifamily. When the largest alt-asset manager in the country says it intends to deploy Claude across its portfolio, the institutional operating bar in our asset class is about to move.

Asset management reporting, leasing workflows, underwriting models, lender packages, resident communications, all of it gets rebuilt over the next 12-24 months at the institutional level.

Operators competing for capital will be judged on their "AI fluency", whether they realize it or not. I'll go so far as to say that eventually, firms without proven AI capabilities won't be able to get through their Equity investor's Investment Committees.

3/ The mid-market is in the addressable target.

This is the part most readers will miss, that the new firm is explicitly designed to serve mid-size companies, not just consortium portfolios.

Goldman's Marc Nachmann framed it as "democratizing access to forward-deployed engineers," and that's not hollow language.

In TMD 065, I wrote about how the largest institutional operators were able to deploy AI-driven pricing tools and negotiate vendor contracts at a scale smaller operators couldn't match on a per-unit basis.

That structural gap is now starting to close. Slowly, but it's closing.

I'm not going to pretend I have all of this figured out. Below are the five questions that I'm thinking about right now, and that I said I'd be sharing today in my LinkedIn post yesterday.

Five questions I'm considering:

1. If Blackstone deploys Claude across its multifamily portfolio first and the mid-market follows 12 to 24 months later, what does that arbitrage window look like for independent operators today?
2. Does this acquisition pattern, (i.e. PE-backed AI services firms targeting mid-sized operators), eventually consolidate the operator side of Multifamily the way RealPage consolidated revenue management?
3. At what point does "we don't have the in-house resources to build with AI" stop being a defensible reason and start becoming a fiduciary problem for Multifamily operators?
4. If the playbook is "applied AI engineers embed alongside operations staff," then is the next job posting at every regional Multifamily operator going to be a Director of Applied AI? And if so, who's training them?
5. The capital backing this new JV firm spans real estate, PE, and sovereign wealth. None of those LPs are paying 2-and-20 to watch their operators fall behind. How long before LP letters start asking GPs about their AI integration roadmap directly?

The point is this: The moat in Multifamily over the next 24 months will not be the AI model anyone uses.

The moat will be how quickly an operator integrates these tools into the boring operational machinery, and how honestly they can address the five questions above.

Hit reply and let me know how you're approaching this today. I'm curious to hear.

The Industry That Was Late, Until It Wasn't

Commercial real estate has always seemed to be a late adopter of technology.

For better or worse, this is simply an observation about the structure of our industry.

Long hold periods, illiquid assets, relationship-driven deal flow, and capital-intensive operations all create gravity that pulls toward "the way we've always done it."

For most of CRE's history, that gravity has been a feature, not a bug.

The operators who endured were the ones who didn't chase every shiny object. In short, discipline beat novelty.

Until it didn't.

Every cycle, a few firms wake up to discover that "the way we've always done it" stopped working five years ago, and they were the last to know.

They underwrote without the new data, leased without the new tools, or reported to LPs in an antiquated manner.

By the time they noticed, the gap was a moat, built by someone else, with them on the wrong side of it.

I think about this as firms becoming victims of their own success.

They spent two decades getting one thing right, and then assumed the same approach would work for the next two.

The skill that built the firm becomes the constraint that limits it.

I see this pattern in our industry constantly, and I think the next 24 months are going to expose a lot of it.

One small example of where the industry is moving sits in due diligence. For most of my career, lease audits, vendor contract audits, and utilities audits have been done the same way.

Essentially, a junior team member opens hundreds of PDFs, types values into a spreadsheet, double-checks the math, and hopes the deadline holds. The work is essential, but it's the kind of work that gets rebuilt first as AI tools mature.

​Rely, this week's TMD sponsor, is one of the firms doing that rebuild for Multifamily.

Their platform converts unstructured documents like leases, vendor contracts, and utility statements into structured, audit-ready data, with every value traceable back to the exact page it came from.

It's trained on Multifamily documents specifically, which matters, because generic AI extraction tools tend to fall apart on the messy reality of a real data room. I'm not telling you to use Rely, but I'm telling you that the operators who replace 80% of their manual diligence work with tools like this in 2026 will be measurably faster, more accurate, and more capital-efficient than the operators who don't.

And, like it or not, that gap will only compound going forward.

Adaptability is one of the most underrated traits in this business. I'm talking about the kind that takes a hard look at the boring operational machinery every couple of years and asks honestly whether the way it's being done still makes sense.

Most of the time, the answer is yes, but sometimes the answer is no. The firms that act on the "no" first will quietly take a growing percentage of market share over the next decade.

We are in one of those moments now.

Stay open-minded. What got us here won't get us there.

The Multifamily Career Skill That Now Matters Most

If you're early or mid-career in Multifamily right now, the most important question isn't whether AI will reshape this industry. I think we can all agree that the answer is an obvious yes.

In TMD 025, I wrote that skill acquisition is the most valuable skill any of us can possess, and that learning how to learn is the meta-skill that makes every other one accessible. I still believe that.

What's changed in the last 18 months is which specific skill, applied on top of that foundation, has the steepest near-term value curve.

I'll tell you what I'm watching when I speak to junior talent across our industry: The ability to cohesively understand the inputs and outputs that generate successful outcomes in the Multifamily business.

My goal has been to learn how to fluently diagnose and communicate each of the various processes that occur across the sector based on where business needs exist at a given point in time.

For example, three years ago I hardly knew what yield-on-cost meant, but since then I've overseen our $400M portfolio, led $300M+ in transactions (refinances + acquisitions), owned our institutional investor relations, and helped raise $50M+ of institutional equity.

Learning the language of Real Estate, and understanding the implications behind how the various inputs generate certain outputs are both invaluable skills.

The Multifamily professionals who can bridge that gap become the most valuable people in the room.

Communicating intelligently, articulately, and effectively all have enduring value.

In fact, communication may become even more valuable as AI absorbs more of the operational work around it.

Weekly Listen

This week's listens feature feature Jon Gray, President and COO of Blackstone, and together they tell the full story behind the Anthropic-Blackstone JV that I overviewed above in Section 1.

The news (4 min): Jon Gray's Squawk on the Street segment on May 4, the morning of the JV announcement. He explains the structure of the deal and what the consortium is trying to build, in his own words.

The strategy (8 min): Gray's Market Views Q2 2026 episode, recorded five days before the JV announcement. He walks through how Blackstone and its portfolio companies are already using AI, and why he sees AI infrastructure as one of the most significant capital deployment opportunities of the next decade.

My recommendation: Watch the first video to understand what was announced, and then watch the second one to understand why it matters.

Wrap Up

That's it for today. I hope you found this edition of The Multifamily Download insightful.

Consider sharing this link to The Multifamily Download with a friend or colleague.

Your feedback is appreciated, so feel free to reply anytime.

Thanks for reading. See you next week!

Forwarded this email? Sign up here.

Join me on LinkedIn | Twitter | Website​