Something quiet but important happened in commercial real estate this month, and most owners missed it because it did not look like a headline. It looked like a series of product launches.
Green Street put out a message that reads almost like a manifesto. Your AI has answers, GreenStreetAI has context. Their argument is that a number without its surrounding market history, news, and forecast is not an insight. It is a starting point. They are right about that. The move they made next is the part owners need to study.
Green Street launched an MCP server. Trepp is connecting AI to trusted CRE intelligence through TreppMCP. MCP is a connector standard that lets an AI model reach into a provider's data on demand. The pitch is compelling: point your AI at institutional-grade intelligence and take your research from days to minutes.
Here is the uncomfortable read for an asset manager. Both of these moves are about who controls the data an AI reasons over. The model is becoming a commodity. Trusted data, context, permissions, and connected workflows are becoming the real asset. The vendors that own proprietary data are racing to make sure that when your AI needs context, it has to come knock on their door.
The model was never the moat
For two years the AI conversation in CRE has been about which model is smartest. That conversation is closing. The performance gap between open-source models and frontier models has narrowed to a sliver, and the cost to reach a given benchmark is falling several times over each year. When capability converges, model selection stops being strategy and becomes procurement. You will pick a model the way you pick a copier.
What does not converge is the data. Your rent roll, your utility consumption, your maintenance history, your tenant behavior, your operating decisions across a hold period. That information is unique to your portfolio, and an AI is only as useful as the data it is allowed to reason over. This is exactly why Green Street and Trepp are moving. They understand that in the AI era the differentiator is not the answer engine. It is the trusted, contextual data feeding it.
They are building connectors to the data they own. The question every owner should ask is simple: who is building the connector to the data you own?
The industry's own leaders are saying it out loud
This is not an OpticWise theory. It is the consensus among the largest owners in the business. In a candid discussion among leaders from Blackstone, Phillips Edison, and Alpha Alternatives hosted by Realcomm, the group landed on a point that should be printed and taped to every investment committee wall. AI is not a new challenge. It is raising the stakes on old ones. Data architecture, connected systems, better workflows, adoption, and execution. Those are the foundation.
One participant put it plainly: they were talking about data architecture and consolidation ten years ago, and if you skipped any of those steps along the way, you are behind now. The AI wave did not create the requirement to own and govern your data. It exposed who skipped the work.
That is the real story of this moment. The owners who treated data as background plumbing are discovering it was structural all along. The ones who built a clean, owned, governed foundation are the ones who can now point any model at their portfolio and get answers that are theirs to keep.
Consolidation is a signal, not noise
Watch where the money is moving. When RealPage acquired Cherre, it was not buying a dashboard. It was buying a data integration and access layer, the plumbing that connects and normalizes property data. The platforms are racing to own the layer that sits between your buildings and the intelligence you want to extract from them.
That should tell an asset manager exactly where value is concentrating. It is not concentrating in the AI model. It is concentrating in the connective tissue: the data plane that holds your operating reality and the trust plane that governs who can touch it and under what permissions. Whoever owns that layer owns the economics of every AI query that runs on top of it.
If that layer sits inside a vendor platform, then your portfolio intelligence becomes a subscription. You will pay to ask questions of your own operating history. And the day you want to switch platforms or bring in a better model, you will discover how much of your intelligence was never portable to begin with.
This is the pattern we see across fragmented buildings, and it is expensive long before AI enters the picture. In one Peak Property Performance® Review of a 400,000 square foot office property, we found roughly $300,000 of redundant fiber backbones running in parallel, each under a different vendor's control, and nobody on the ownership side could fully explain them. That is what happens when the data & digital infrastructure belongs to your vendors instead of to you. Now imagine that same fragmentation governing which AI can see your data and which cannot.
If you don't own the data plane, you are renting your own intelligence back
Here is the reframe that matters. If you don't own your data & digital infrastructure, your vendors do. In the AI era, that sentence stops being about wiring and starts being about intelligence itself. The vendor who controls the data plane controls what your AI can know, what it can act on, and what it costs you to find out.
The MCP announcements from Green Street and Trepp are genuinely useful products, and I am not telling anyone to avoid them. Market comps and securitized loan intelligence are exactly the kind of external context an AI should be able to reach. The distinction is this: those are external data sources. Your operating data is different. It is the proprietary record of how your specific assets perform, and it is the one input a competitor cannot replicate. That is the data plane you cannot afford to rent.
When owners let their own operating data get trapped inside vendor platforms, they end up with automation without governance and answers without ownership. The intelligence compounds for the vendor, not for the portfolio.
Build the owned foundation before you connect the models
The way out is not more tools. It is a foundation you control, and then any model or connector you want plugs into it under your rules. That sequence is the whole point of the PPP 5C™ plan, and it maps directly to this moment.
Clarify. Start with a Review that defines your success metrics, maps who actually owns and can access your operating data, and documents what is trustworthy and portable today. Most owners discover the honest answer is: less than they assumed.
Connect. Establish secure, owner-controlled connectivity that repeats property to property, so your buildings feed one owned foundation rather than a dozen vendor silos.
Collect. Capture and normalize your operating data into a consistent model you can reuse, not a format locked to whoever collected it.
Coordinate. Govern identity, access, privacy, lineage, retention, and the rules of use. This is the trust plane. It is what lets you decide which model, which vendor, and which connector gets to reason over your data and under what permissions.
Control. Now enable any decision engine, any internal analytics, and any AI model to act under your permissions. This is where Property Brain™ becomes real: a vendor and LLM-agnostic intelligence layer for a single asset. Standardize it across the portfolio and Property Brain™ becomes Portfolio Brain™, where intelligence compounds across buildings instead of restarting at every address.
That is the structural difference between the OpticWise approach and the connector race. Green Street and Trepp are building doors into their data. We help you own the house. When your data plane and trust plane belong to you, MCP connectors, frontier models, and whatever comes next all become tools you point at your foundation, not landlords you rent your own intelligence from.
The asset managers who understand this will spend the next twelve months building an owned, governed data foundation while their peers are signing more subscriptions. When AI capability fully commoditizes, and it will, the portfolios that own their data plane will be the ones turning intelligence into capitalized value. Every dollar of NOI you protect or grow through better operating decisions is worth roughly fifteen to twenty-five dollars of asset value at typical cap rates. That math does not run on a rented data plane.
Start with a Review of one property. Establish Property Brain™ on a foundation you own. Prove the portability by pointing a decision engine or a model at it. Then productize the standard and scale it into Portfolio Brain™. That is how you make sure the AI era compounds value for you instead of for the platform holding your data.
Own your data & digital infrastructure. Operate with strategic foresight. Build for the long game.
