Inspiren

The model is the engine of the resident safety module, and the ecosystem carries value beside it. On the resident safety page, 'AI risk detection flags unsafe movements or behaviors' from the AUGi room devices, which the company calls physical AI and which capture 'AI-powered blurred' images for alerts, a privacy-protected Live View and event replay; Inspiren HQ applies AI to data from all devices to produce clinical insights.

The rest of what a community buys stands without the model: an emergency call system of help buttons, pendants and two-way voice, staff beacons that log time spent in each apartment, heat maps and staffing dashboards, care level reviews, and a clinician-led success team assigned to each community.

The staff efficiency page says staff beacons 'Ensure every care moment is logged automatically', the same page promises 'every minute of care documented', and the home page reports '100% of care documented'. Those are completeness claims about the system's own record of care, and a review step cannot show what was never logged; the logged care time is what drives care level and billing reviews.

The oversight structure around resident alerts is otherwise described: AI risk detection alerts staff with a privacy-protected visual and two-way voice, a staff member claims the alert so ownership is clear, incidents can be replayed, and Inspiren's clinical team reviews flagged care plans with community leaders before care levels change. The company publishes no alert thresholds or false alarm rates, and does not say what happens when an alert is missed. The claim of 'complete coverage across the entire community' and '100% coverage in all rooms' is a coverage statement that sets the scope this oversight has to answer for.

The sensing concept is described and the detector is not characterized. The AUGi page describes 'AI-powered blurred image capture' for resident privacy, with a privacy-protected Live View and event replay; the resident safety page says AI risk detection flags unsafe movements or behaviors; and the September 2026 funding release calls the approach physical AI combining ambient awareness with longitudinal resident data.

No model or method is named, no training data or development population is described (mobility aids, posture, lighting and room layouts matter in memory care), no detection performance is published (sensitivity, false alarm rate, time to alert), and no versioning or update practice is stated.

The architecture is described and no party in it is named. Room devices capture blurred images and signals from beacons, pendants and help buttons, and 'All data flows into Inspiren HQ and Mobile App'. The privacy policy (last modified 15 January 2026) says information may be shared with vendors and service providers, analytics providers such as Google Analytics, advertising partners and affiliates, and it addresses the service generally rather than resident data. No model provider, hosting provider or subprocessor list is named, and nothing says whether detection runs on the device or in the cloud.

Named deployments with dated outcome figures, reported by the vendor and short of independent evaluation. Inspiren's March 2026 release reports a 10 month study at Solera Senior Living's Lumina Las Vegas memory care community (January to October 2025): 48 percent fewer falls, 54 percent fewer emergency room visits and hospitalizations, 50 percent faster staff response, and occupancy up 24 percent to 93 percent.

Its September 2026 release adds Aegis Living (residents who opted in stayed 34 percent longer, with a 22 percent lower fall rate and 24 percent lower injury rate) and Clearwater Living (in the first three months, falls with injury down 63 percent and post-fall emergency visits down 73 percent). No resident counts, baseline periods or comparison groups are given beyond the Aegis opt-in comparison, which may reflect who opts in; Solera also partners with Inspiren in a care model alliance. The home page's 68 and 83 percent fall figures name no community. No detection accuracy figure and no peer reviewed or independent evaluation is published.

A categorical design commitment is published without retention or training terms. The AUGi page says the device 'Maintains resident privacy with AI-powered blurred image capture', with Live View and replay described as privacy-protected, so what is captured in a bedroom or bathroom is a blurred image rather than footage; the September 2026 release refers to residents who opted into Inspiren, which indicates participation is by choice.

The same page claims 'HIPAA-compliant safeguards at every layer, from encrypted data handling to strict access controls'. Nothing states how long images, replays, movement data and care time logs are kept, whether resident data trains or improves the detection models, or whether data is de-identified for any other use; the privacy policy (15 January 2026) does not address resident data. Safety engineering, meaning how a missed or false alert is investigated and handled as a safety event, is not published.

Compliance is claimed and no document supports it. The AUGi page says the product is 'Built with HIPAA-compliant safeguards at every layer', and the only privacy policy (last modified 15 January 2026) covers the service generally with contact, usage, analytics and advertising data and says nothing about resident health information or business associate status. No agreement, role statement or breach terms are published.

Whether federal health privacy law reaches this data at all depends on the operator: many assisted living and memory care communities are not covered entities, in which case no business associate relationship exists and nothing federal governs the data.

Controls are asserted with nothing independent behind them. The AUGi page claims encrypted data handling and strict access controls; no SOC 2 report, HITRUST or ISO/IEC 27001 certification, penetration test or trust center is named on the home, product, hardware, software or legal pages. The holding is continuous sensing in residents' rooms and bathrooms, blurred imagery, replays, and logs of every staff visit, for a largely cognitively impaired population.

No clearance is claimed and no regulatory position is stated. The product detects falls and risky movement and alerts staff, and the care planning and staff efficiency pages say 'AI-powered technology flags changes in condition and guides care decisions'; the home and release pages market fall, injury and hospitalization reductions.

What sits behind those outputs is reviewable by the people acting on them: alerts arrive with a privacy-protected visual and two-way voice, incidents can be replayed, and care plan flags rest on logged care time that Inspiren's clinical team reviews with community leaders. The tension is that condition change detection and fall outcome claims are clinical, and nothing says how the company classifies them.

There is no published account of how the detection models are governed or tested. The site says a clinician led success team works with each community and that 'experienced nurses and clinical experts ... ensure our technology works hand-in-hand with human care', which describes human support around the product rather than any evaluation of model behavior.

No detection results are published by mobility aid use, posture, body size, lighting or time of night, although residents most likely to fall are the ones whose movement is hardest to track, and falls concentrate at night.

Mechanisms exist that let someone check an output, and nothing stands behind it. Event Replay lets a team 'Replay events like falls or unusual activity to understand what happened', alerts arrive with a privacy-protected visual so staff can see what triggered them, alert claiming records who responded, and care plan flags are reviewed with Inspiren's clinical team before care levels change.

The terms of service provide the service 'as is', cap Inspiren's liability at the greater of $50 or three months' fees, and run indemnity from the user to Inspiren. No detection accuracy or false alarm rate is published, and no route is described for a resident or family when a fall was missed or a care level was raised on the system's data.

Named record systems with read access, and outbound writing claimed without a named system. August Health's own integration page (August 2026) says resident names and room assignments flow from August Health into Inspiren automatically as a first, census-level phase; an Extended Care Professional integration announced in August 2026 was described as syncing census, transfers, discharges and level of care updates into Inspiren, though that page no longer loads.

Inspiren's own pages say care time is auto-tracked 'and sync[ed] to the EHR' and that it 'Connects with leading EHRs and data systems', naming none, and a customer quote describes integration with an operator's business intelligence platform. What leaves Inspiren for the resident record is not documented for any named system, so falls and care findings may live only in Inspiren HQ.

Hardware in every room feeding a single hosted platform, with location implied rather than committed. AUGi devices, sensors, pendants, help buttons and staff beacons are installed across a community, and 'All data flows into Inspiren HQ and Mobile App'. No hosting provider, region, tenancy model or on premise option is stated, nothing says whether detection runs on the device or after data leaves the room, and no retention or return of data at contract end is described.

No price is published, and the shape of the sale can be established: 'a powerful mix of hardware, software, and expert clinical support' per community, sold through a demo. The economic case is published in the operator's terms rather than the price: '$840 increase in annual revenue per unit', '$70/m NOI increase per resident', '31% of care plans adjusted to correct levels' and, in the September 2026 release, longer resident stays. Care level determines a resident's monthly fee, so a system that finds care levels set too low raises residents' bills; nothing published says how often care levels were adjusted down.

Coverage is named and bounded to senior living communities, assisted living and memory care, across resident safety, emergency call, care planning and staffing, including bathrooms and common areas; no acute, clinical diagnostic or in-home care is claimed.

Validation sits behind part of it: dated outcome figures come from memory care (Solera's Lumina Las Vegas, 2025) and from Aegis Living and Clearwater Living (September 2026 release), while assisted living use rests on operator quotes (Charter Senior Living, Beacon Communities, Arrow Senior Living). The product is sold to facilities, not to home care agencies.