Insurance companies are using drones, satellites, and AI to scan millions of roofs — without homeowner knowledge or consent. The AI flags your roof. You get a letter demanding a $20,000 replacement or loss of coverage. And in most states, you have no legal right to see the images used against you. Here's what's happening and how to protect yourself.
Get an Independent Roof Inspection →Insurance companies have used satellite imagery for decades. What changed in 2023-2026 is the combination of three technologies deployed at scale: high-resolution aerial imagery from drones, fixed-wing aircraft, and satellites; AI models trained on thousands of images to detect roof risk indicators; and companies with coverage of 99.6% of the U.S. population providing this data as a service to insurers.
Here's how the process works. An aerial imaging company — CAPE Analytics, Nearmap, EagleView, and others — captures high-resolution imagery of your property. That imagery is fed into AI models trained to identify risk indicators. The AI produces a risk score. Your insurer receives that score and uses it to make underwriting decisions — renew, non-renew, demand repairs, or raise your premium. You receive a letter. You had no idea any of this was happening.
State Farm, Allstate, and Farmers are among the carriers using this approach. State Farm contracts third-party aerial imaging companies rather than flying drones directly — a legal distinction that changes nothing practically. Your roof gets assessed. You don't find out until the letter arrives.
A homeowner who had lived in her home since 1993 received a notice from State Farm demanding she replace her roof at an estimated cost of $20,000 or risk losing coverage. No one had set foot on her property. State Farm had used aerial imagery analyzed by AI. Her response: "My initial thought was it's a mistake. They've got the wrong house because there's nothing wrong with my roof." Consumer advocates confirmed the AI "makes a conclusion that's wrong about what it sees" in a meaningful percentage of cases.
AI roof inspection systems are trained on large datasets of roof images paired with known outcomes — claims, replacements, damage events. They learn to identify visual patterns that correlate with elevated risk. Understanding what they flag helps you understand what triggers a non-renewal or repair demand.
Color variation and dark patches on asphalt shingles indicating loss of protective granule layer. Often flagged as the roof approaching end of useful life.
Green or dark streaking on shingles. AI flags this as a moisture retention and deterioration risk. Common on older coastal roofs in humid climates.
Visible gaps in shingle coverage, often after storm events. High-confidence flag — missing shingles are usually clearly visible in aerial imagery.
Standing water on flat or low-slope roofs. Indicates drainage failure and is a significant flag for commercial properties and flat-roof residential homes.
Tree branches within a measured distance of the roof surface. Flagged as debris impact risk and a source of ongoing leaf accumulation and moisture retention.
Visible deformation of roof planes. Usually a high-confidence flag when present, indicating deck or structural issues below the surface.
AI inspection systems produce errors at a rate that consumer advocates and state regulators consider significant. The errors are systematic — they arise from the limitations of aerial photography and the AI's training data, not random chance.
Dark solar panels on a light roof surface have been misidentified by AI as missing or damaged shingles. Homeowners with solar installations are disproportionately affected.
Moss growing on a tree adjacent to the property line has been counted against the homeowner's roof shingles due to aerial perspective overlap.
At least one documented case in Texas of a homeowner receiving a cancellation notice based on aerial images of a completely different house — addressing error in the AI data.
Depending on time of day and sun angle when the aerial image was captured, shadows from chimneys, vents, and overhanging features have been flagged as dark patches indicating damage.
Leaf accumulations, pine needles, and seasonal debris on the roof surface have been misread as gaps in shingle coverage or signs of deterioration.
Different shingle batches installed at different times — common on repaired sections — show color variation that AI systems misread as accelerated granule loss.
— Consumer advocate, United Policyholders, March 2026. Consumer advocates and state insurance departments have confirmed that AI roof inspection errors are not isolated incidents. The Illinois homeowner who had a replacement demand dropped after filing a state insurance commissioner complaint "is not an isolated case" — it is a pattern that state regulators are beginning to address legislatively.
The current legal framework in most states offers homeowners very limited protection against aerial AI inspections.
California — passed legislation requiring 30-day advance notice before using aerial imagery for coverage decisions.
Pennsylvania — mandates 60-day notice periods before acting on aerial inspection data.
Massachusetts — proposed legislation (not yet passed as of mid-2026) that would give homeowners the right to see images, know when they were taken, and have a formal appeals process.
In all other states — including Florida, Texas, Louisiana, and the other Gulf and Atlantic coastal states — insurers can conduct aerial AI inspections with no notice required, and homeowners have no statutory right to the underlying data.
Most states allow homeowners to appeal coverage decisions with evidence submission. However, the practical challenge is significant: you are arguing against an AI system's conclusion using manual evidence — photos, contractor reports — while the AI's imagery may be months old, taken from an angle that captured an anomaly, or simply wrong about your property. Filing the appeal with a licensed contractor's written report is your strongest counter-evidence.
A licensed contractor's written inspection report documenting your roof's actual condition is the strongest counter-evidence in an appeal. Our free inspections connect you with licensed local roofers who provide detailed written documentation.
Get a Free Documented Roof Inspection →If you receive a non-renewal notice or roof repair demand based on an AI inspection you didn't know about, here is the step-by-step response that gives you the best chance of a favorable outcome.
Contact your insurer right away and request a 30-60 day extension while you gather evidence. Most insurers will grant this — they prefer to retain policyholders who make repairs rather than process cancellations. Get the extension in writing.
Ask your insurer in writing for the aerial images, AI report, date of inspection, and the specific deficiencies they identified. They may decline — but the request creates a paper trail and in some cases insurers do provide the data, especially if litigation is suggested.
Hire a licensed contractor for a full roof inspection and get a detailed written report documenting current condition, material type, estimated age, and absence of the specific deficiencies cited. Take date-stamped photos from ground level covering every roof plane. This is your counter-evidence.
Submit the contractor's report and your photos as a formal appeal of the coverage decision. Request that a human adjuster review the evidence in conjunction with the aerial analysis. Be specific about any documented AI error types — solar panels, adjacent tree coverage, normal color variation — if applicable.
This step is more effective than most homeowners realize. One Illinois homeowner had a replacement demand dropped after filing a commissioner complaint. The threat of regulatory scrutiny often prompts insurers to send a human adjuster for an in-person assessment. In coastal states, the insurance commissioner's office handles thousands of homeowner complaints and tracks patterns — your complaint contributes to regulatory pressure for reform.
If your appeal is denied, work with an independent insurance agent who represents multiple carriers. Some insurers have more flexible underwriting for older roofs. In states with active wind pools (Florida Citizens, Texas TWIA, Louisiana Citizens), you have a last-resort option even if private carriers non-renew.
The most effective protection against an AI-generated non-renewal is having recent, dated, professional documentation of your roof's condition already on file before any letter arrives. Annual inspection reports create a documented record that pre-dates any insurer's aerial analysis — making it harder for an insurer to claim conditions existed at an earlier date than your photos show. Think of it as a pre-inspection baseline that protects you the same way pre-storm photos protect your storm claim.
Beyond underwriting decisions, drone and AI technology is now central to how insurers process storm damage claims. Understanding this changes how you should approach a claim after a hurricane or major storm.
After a named storm, insurers compare pre-storm aerial imagery against post-storm images to assess damage. This comparison drives their initial damage estimate. If your roof had pre-existing conditions that are visible in the pre-storm imagery — granule loss, algae, overhanging trees — the insurer may use these to argue that some post-storm damage was pre-existing or attributable to deferred maintenance rather than the storm.
Drone and AI-based claim assessment allows insurers to process claims significantly faster after a major storm event. However, faster also means less nuanced. Subtle damage — broken seal strips, minor flashing displacement, underlayment tears — that requires an inspector on the roof to detect may be missed in an aerial-only assessment. The claim is processed, paid at the aerial estimate, and closed — without capturing the full scope of damage.
Having a licensed contractor's written inspection report documenting all damage — including the non-aerial-visible types — before the adjuster's formal assessment gives you the documentation to supplement the aerial estimate. Contractors who work regularly with storm claims know how to document damage in the format adjusters and carriers recognize. This is why many roofing contractors now attend adjuster inspections alongside the homeowner.
Insurers capture post-storm imagery as quickly as possible after a named storm. If you wait weeks to inspect your roof, the aerial comparison they're working from was already taken. Subsequent rain events, debris, and continuing weathering can change the roof's condition, potentially giving the insurer grounds to attribute some damage to post-storm weather rather than the original event. Inspect within 48-72 hours of a storm — ideally before the insurer's aerial assessment is even processed.
The 7 damage types that aerial AI will miss — and why you need a contractor on the roof.
What photos, videos, and written records you need to supplement an insurer's aerial assessment.
Why having your own inspector there when the adjuster arrives can mean thousands more on your claim.
How AI inspections combine with age-based ACV coverage to create the worst possible scenario for older roofs.
How a FORTIFIED upgrade eliminates most of what AI flags — and comes with free grant money in 6 states.
When to hire your own adjuster to counter an insurer's AI-driven claim assessment.