XHome Survey and Visionary AI Engine: What Future Proof Is Building

XHome Survey and Visionary AI Engine

XHome Survey and the Visionary AI Engine are Future Proof Property Intelligence's property-specific inspection and remediation-planning products.

The company describes XHome Survey as a guided smartphone scan of a home and site. Visionary is described as the analysis layer that turns captured and external data into vulnerability findings, itemized construction recommendations, material quantities, cost information, and a record for rescoring after work is completed.

The intended workflow is clear. The current operating status is not.

Future Proof's February 2026 SEC filing said Visionary had been released in July 2025 and projected XHome Survey for the first quarter of 2026. A company article published during the current cycle said XHome Survey was "to be released in 2026." Current public product pages describe the capabilities but do not provide a public app-store listing, general-availability announcement, service-level description, pricing, or independent validation record.

The products should therefore be described as in development or limited availability until a current authoritative source establishes otherwise.

Intended product chain

flowchart LR
    A["Guided phone capture"] --> B["Property and site model"]
    B --> C["Vulnerability analysis"]
    C --> D["Remediation scope and budget"]
    D --> E["Contractor or DIY execution"]
    E --> F["Verification and rescore"]
    F --> G["Insurer review"]

The company describes this full chain. Public evidence does not establish that every step is generally available or accepted by insurers.

XHome Survey

Future Proof says XHome Survey guides the user through an exterior property scan. The target is broader than the house alone. The Visionary product page refers to the house, lawn, garage, outbuildings, and other site elements.

A guided capture can improve consistency. It can prompt a user to photograph roof edges, openings, attachments, vegetation, drainage, utilities, and accessory structures. It can also preserve a dated condition record.

The scan has physical limits. A phone cannot reliably inspect every roof surface, concealed connection, foundation condition, wall assembly, drainage elevation, or code issue. Image quality, weather, access, lighting, user behavior, geolocation, and model inference can affect the record.

A responsible output should identify what was directly observed, what came from another dataset, what was inferred, and what requires professional investigation.

Visionary AI Engine

Future Proof describes Visionary as the analysis engine behind property dossiers, construction recommendations, and before-and-after risk scores. The current product page says it will provide structural-vulnerability findings, itemized recommendations, material lists, and an underwriter property-rescoring report.

The company's methodology page says its broader process combines AI with human review and that it intends to validate completed work and issue a certificate for an insurance carrier.

Those statements establish product intent. They do not establish that a licensed engineer designed a particular plan, that a local jurisdiction accepts the output, or that an insurer recognizes the certificate.

Future Proof's offering statement identifies inaccurate AI-generated risk scores and retrofit recommendations as a material risk. It specifically notes possible under-mitigation, property damage, denied claims, rescinded discounts, liability, and regulatory review.

That disclosure defines the product's central safety requirement. The system must keep AI assistance separate from professional authority.

What a credible property model needs

A credible model needs a stable property identity, current location, applicable hazard, building and site observations, provenance for external data, explicit assumptions, and uncertainty.

The NAIC's catastrophe-model overview explains that expected loss depends on hazard, vulnerability, exposure, and policy terms. Construction, occupancy, height, location, insured value, limits, and deductibles can all affect the result.

A product-specific risk score should therefore not be presented as the insurer's score. Different carriers and model vendors may use different data, event sets, damage functions, and underwriting rules.

The most valuable output may be the evidence behind the score: dated observations, verified attributes, recognized standards, completed work, and clear limits.

From recommendation to construction

The product page describes precise improvement specifications and a complete construction roadmap. Those phrases should be tested against the responsibility chain.

Some recommendations can be educational. Others may constitute engineering, architecture, contracting, surveying, code, or insurance activity under local law. The product should identify when a qualified professional must review, design, stamp, permit, or inspect the work.

The construction scope also needs interfaces. A roof-cover recommendation can be incomplete without deck attachment, water barrier, flashing, openings, and edge details. A wildfire recommendation can fail if it ignores near-home fuels. A flood recommendation can be impermissible for a residential building in a regulated flood zone.

FEMA's floodproofing guidance is a good example of why a national model cannot issue a generic measure without local context.

Cost estimates and product lists

Future Proof says the system will generate itemized costs and material lists for contractor or do-it-yourself use.

An estimate can help establish order of magnitude and compare options. A buildable budget must also account for professional services, permits, testing, demolition, temporary protection, labor, disposal, concealed conditions, inspection, site restoration, and contingency.

The company says plans may link products from manufacturers and retailers. That creates a disclosure need. A recommendation should distinguish technical suitability from a paid placement, affiliate relationship, lead fee, or sponsorship.

The user should also be able to substitute an equivalent approved product when the standard and design allow it, rather than treating a commercial link as the engineering specification.

The insurer-facing rescore

Future Proof's most ambitious product claim is the connection to underwriting.

A before-and-after record can be useful if it shows a verified change relevant to expected loss. An insurer still must decide whether the data fits its model, whether the measure is recognized, whether a filed discount applies, and whether portfolio conditions permit coverage.

The SEC filing says carriers may require years of actuarial validation and may prefer incumbent or internal solutions. That is a direct acknowledgment that the Future Proof rescore cannot bind the insurance decision.

Existing programs show what acceptance can look like. IBHS's FORTIFIED program uses named construction standards and evaluation for severe wind and rain. California's Safer from Wildfires framework ties specified measures to discounts under state rules.

The lesson is not that Future Proof should copy one certificate. It is that the insurer-facing output needs a defined peril, accepted standard, accountable evaluator, durable documentation, and loss evidence.

Availability and verification questions

Before representing the products as available, confirm whether XHome Survey can be accessed by the general public, which geography and property types are supported, what professional review is included, what a report costs, which inputs are collected, how data is retained, and which insurers or programs accept the output.

Also confirm whether "Visionary released July 2025" means an internal engine, a partner deployment, a limited beta, or a generally available service. The public SEC and website records use release language differently.

This status ambiguity is not unusual for an early-stage platform. It should remain explicit because the product affects property, safety, privacy, and financial decisions.

Venture Step context

Ben Gilliland describes the scan-to-remediation system in E112 as an answer to a fragmented process. The episode is valuable because it focuses on the handoff from risk information to physical work.

Read From Risk Score to Retrofit for the complete decision framework. How Address-Level Risk Data Could Change Home Insurance explains why a more detailed property record still needs model validation, regulatory treatment, and insurer adoption.

Verification and disclosure

This profile was checked on July 27, 2026 against Future Proof's product, methodology, and company pages, its February 2026 SEC offering statement, and independent insurance and mitigation sources.

Product status, pricing, geography, professional services, insurer acceptance, data practices, and technical methods require refresh. Nothing here is an endorsement or construction, engineering, insurance, legal, investment, or financial advice. AI assisted with research organization and drafting; source boundaries and final editorial decisions remain Dalton Anderson's.