AI-SMART-BUILDINGS.COM · OPERATOR BRIEF · 2026-08-04

Every major smart-building vendor says AI is central to its strategy. Independent research finds that almost none of them lets you verify it from their financials. Here is what that means for a buyer — and the questions that cut through it.

If you are evaluating an AI-enabled building platform in 2026, you will hear the same message from nearly every incumbent: artificial intelligence is core to what we do. The harder question — how would I independently verify that? — has a more revealing answer.

The independent smart-building research firm Memoori assessed nine of the largest incumbents (Johnson Controls, Honeywell, Siemens, Schneider Electric, Trane Technologies, Carrier, ABB, Delta, and Acuity) in its 2026 study. Its finding on disclosure was blunt: no incumbent breaks out AI or platform revenue as a reporting segment, even though all position AI as central to strategy (Memoori 2026). Buyers get segment-level and increasingly data-center revenue — for example, Honeywell's Building Automation segment at roughly $6.5 billion (FY2024) — but no AI-attributable line that would let a customer or investor gauge real traction.

This is not an accusation of bad faith. It is a structural fact about how large diversified manufacturers report. But it does mean the burden of verification sits with you, the buyer — and that generic "AI everywhere" marketing is the weakest possible basis for a purchase decision.

Where the evidence is actually strong

Transparency is not uniformly absent. Two data points stand out because they are specific, at scale, and third-party-verified:

The lesson is not a ranking of one vendor over another. It is that specific, scaled, independently verified numbers exist — so you are entitled to ask for them, and to treat their absence as a signal.

Five questions that cut through the AI claims

Rather than rank vendors on marketing, put every candidate through the same five checks. Ask for a written answer to each:

  1. What does the AI actually decide or control? A system that autonomously adjusts equipment is a different product from a dashboard that recommends. Make the vendor name the specific decisions the model makes.
  2. Can you show a reference deployment with numbers? Buildings under management, energy or cost saved, and the baseline those savings are measured against.
  3. How are savings verified? A vendor dashboard is not verification. Ask whether results are measured with a recognized measurement-and-verification discipline such as IPMVP (Option A/B/C), and by whom.
  4. What happens to your data? Cloud vs. edge architecture, data ownership, and lock-in. A clear answer here is itself a transparency signal.
  5. Which parts of this product are getting new investment this year? With incumbent capital increasingly flowing to data centers, confirm the commercial-building features you need are still on the roadmap.

The buyer's takeaway

"AI is central to our strategy" is a starting point, not evidence. In 2026 the market's own research shows that AI-specific results are rarely disclosed at the financial level — so the operators who buy well are the ones who demand specific, verified, IPMVP-grade numbers and test them against their own building data. That verification-first posture is exactly what an independent portal like AI Smart Buildings is built to support.

Vendor figures cited to Memoori's 2026 incumbent research and each company's public reporting; see links above. Independent analysis, not a certification, endorsement, or investment advice.

The Intelligent Building BriefOperator-grade CRE-AI intelligence, weekly. Subscribe free →