What happened
The U.S. General Services Administration (GSA) and the Department of Energy tested building software at the Foley Federal Building and U.S. Courthouse in Las Vegas. The National Laboratory of the Rockies (NLR) and Slipstream measured the results for them and reported in September 2026. The software, Nantum OS from Nantum AI, adjusts heating, ventilation and air conditioning through the building automation system. The report says that it uses machine learning to predict peak demand and to recommend start-up and ramp-down times. Comparing weather-normalized use with the software off and on, the evaluators found that the building used 4.2% less electricity and 5.0% less gas. In three simulated demand response events, demand fell by 8% on average. Continuous demand management produced no measurable reduction. Hot calls from occupants rose 154% and cold calls 50%, and staff switched the software off on the last day. Annual savings of $6,878.79 fell short of yearly fees of $23,044.56, so the software never paid back.
Sources
-
Evaluation report, by an evaluator that the makers or users commissioned.
Sean Pachuta and Alicen Kandt (National Laboratory of the Rockies) and Scott Schuetter (Slipstream). Field Validation of a Grid-Interactive Efficient Building Software Solution. NLR/TP-5R00-89911, September 2026. Funded by the U.S. General Services Administration and the U.S. Department of Energy.
The GEB software used this occupancy data, along with thermal modeling and machine learning algorithms, to recommend start-up times, midday ramps, and end-of-day ramp-downs to building operators.
With time-series data from the BAS to inform the GEB software, whole-building energy savings of 4.3% were realized during the evaluation period.
The GEB software was not found to be cost-effective in this application.
Details
AI system
- Workflow
-
The software connected to the building's Niagara automation system and pulled real-time data. Occupancy counters at each entrance fed the software's start-up and ramp-down recommendations. The software ran demand response in stages during three simulated events.
Money
- Kind of value
-
Annual energy cost savings at the Foley Federal Building at its blended utility rate, before the software's yearly fee of $23,044.56
- Amount
-
Estimate: $6,878.79
- Recipients and payers
-
U.S. General Services Administration
Follow-up
- Conditions for replication
-
Occupants must accept wider temperature swings, which drew more hot and cold calls here. Utility rates must be high enough for savings to cover installation costs and yearly fees.
Entry details
- Sector
-
Energy (building efficiency).
- Purpose
-
Nantum AI built Nantum OS to bring a building's data into one platform and to cut energy use and peak demand through supervisory control.
Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, beneficiaries, reach, distribution and access, duration, every measurement question, currency and prices, period and unit, realization, gross or net, included costs, comparison, method, return and payback, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.
Cite this record
Peter H. Vartanian. (2026). Nantum OS trims a Las Vegas courthouse's energy use by 4.3% (AIBD-0857). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0857.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)