Ford Patents a UBI Rating Engine for a Data Pipeline It Shut Off

Ford Global Technologies, LLC won US Patent 12,694,456 on July 28, 2026, for a system that runs high-dimensional Bayesian optimization against an archive of vehicle signals and their insurance effects, automatically builds weighted aggregation functions from the result, and pushes those functions into vehicles so they compute a composite usage-based insurance score on board rather than transmitting raw driving data. Ford filed the application on February 12, 2024, the same year it wound down the consumer telematics-sharing program the method was designed to run.

That timing is the story. Ford is not a telematics vendor filing patents to defend a live data business; it is an automaker that spent 2024 walking away from third-party data sharing while simultaneously patenting the mathematics that would let it, or a licensee, price usage-based insurance from vehicle signals without ever handing a carrier the raw feed. The grant arrives five months after the Federal Trade Commission finalized a twenty-year consent order against General Motors and OnStar for selling that exact kind of data to brokers without consent, and three months after California extracted its largest privacy settlement to date over the same conduct. Ford’s patent, whatever its intended use, describes an architecture that sidesteps the specific legal theory that caught its rival.

What Claim 1 Actually Builds

The patent’s abstract and independent claims describe a two-stage system. First, a server-side process performs multi-dimensional Bayesian optimization over a data archive containing telematics signals and their observed effect on insurance outcomes, iterating “until a convergence criterion is met and/or until an iteration budget is exhausted” (US Patent 12,694,456, USPTO, July 2026). Each iteration trains a surrogate model, evaluates an acquisition function, and tests a new weighted grouping of signals against the UBI rating model, generating a fresh input-output pair for the next round. Claim 4 specifies that each resulting aggregation function “applies respective weights to each of the plurality of signals,” with the weight representing that signal’s relative importance to the outcome being predicted; claim 5 adds that weights falling below a threshold are set to zero, dropping low-relevance signals from the aggregate entirely.

Actuaries will recognize the shape of this even if the vocabulary is unfamiliar. Selecting which telematics variables belong in a rating algorithm and how much weight each deserves is a factor-selection exercise every UBI pricing team already runs by hand or through a generalized linear model, typically on a monitored cadence tied to a filed rate revision. Ford’s claims automate that selection as a closed loop that runs against Ford’s own archive, reoptimizes without a scheduled model build, and then deploys the resulting weighting scheme directly to the vehicle’s firmware. The second stage is the part with the sharper competitive edge: the vehicle, not the insurer’s ingestion pipeline, computes the weighted composite locally and transmits that aggregate rather than the underlying signal stream, a design the patent frames as reducing vehicle-to-cloud data transfer. Whoever controls that firmware controls which raw variables ever leave the car, and in what combined form.

A Patent Filed as the Data Pipeline Was Shutting Down

Before late 2024, a Ford owner who wanted usage-based insurance had to opt in twice, once with an insurer and again on the vehicle’s own screen, before Ford would relay telematics data to that carrier. Ford discontinued that dual-consent sharing arrangement in 2024, telling customers that “privacy is central to our relationship with our customer and we take it extremely seriously” and that insurers never received connected-vehicle data without explicit consent (Ford Authority, March 2024). The move followed a wave of reporting, prompted in part by Senator Ron Wyden’s investigation, that General Motors had been piping OnStar Smart Driver location and driving-behavior data to brokers who resold it to insurers, in some cases collecting geolocation as often as every three seconds.

That GM story has since resolved into hard regulatory outcomes. The FTC finalized its consent order against GM and OnStar on January 14, 2026, by a 2-0 commission vote, barring GM from selling geolocation or driver-behavior data to any consumer reporting agency for five years and requiring affirmative, express consumer consent before collecting, using, or disclosing connected-vehicle data for the full twenty-year life of the order (Federal Trade Commission, January 2026). California followed in May with a $12.75 million settlement, the largest fine yet levied under the state’s privacy law, over the same underlying conduct: selling driving and location data from hundreds of thousands of California drivers to data brokers Verisk Analytics and LexisNexis Risk Solutions between 2020 and 2024, netting GM roughly $20 million nationwide across that window (California Department of Justice, May 2026). “General Motors sold the data of California drivers without their knowledge or consent and despite numerous statements reassuring drivers that it would not do so,” California Attorney General Rob Bonta said. “This trove of information included precise and personal location data that could identify the everyday habits and movements of Californians” (California Department of Justice, May 2026). The settlement was brought explicitly on data-minimization and purpose-limitation grounds, the principle that a company should not collect or retain more personal data than a stated purpose requires.

Ford’s patented architecture happens to be built around exactly that principle. A vehicle that computes a weighted composite score on board and transmits only the aggregate, never the raw geolocation trace, structurally cannot be the subject of a complaint that it sold second-by-second location pings to a data broker, because that raw stream never leaves the car in the first place. Whether Ford’s engineers designed claim 4 and claim 5 with GM’s looming enforcement exposure in mind or arrived at data minimization for bandwidth-cost reasons that turned out to double as a legal shield, the effect is the same: a UBI architecture that inoculates against the specific theory of liability that produced the industry’s two largest connected-car privacy penalties to date.

From Selling Signal to Owning the Rating Math

The GM settlements describe one business model: an automaker as a raw-data supplier to LexisNexis and Verisk, the two firms that dominate third-party telematics scoring and that GM's own data fed (NAIC Center for Insurance Policy and Research). Ford's 2026 grant, combined with what the automaker has actually built since discontinuing consumer data-sharing, points toward a different model: owning the rating output rather than selling the rating input. Ford Pro Insure, the commercial-fleet insurance product Ford launched through a managing general agent arrangement with Pie Insurance, runs through The American Road Insurance Company, which Ford Pro describes as the first automaker-owned carrier offering a commercial auto insurance product. The program is live in Arizona, Illinois, Indiana, Tennessee, and Wisconsin, with Ford Pro stating its intent to expand nationally over the next several years, and it is fed by Ford Pro E-Telematics, offered as a complimentary one-year trial on eligible 2024, 2025, and 2026 model-year vehicles.

Patent 12,694,456 is not written as fleet-specific; its claims cover usage-based insurance rating generally, which means the aggregation-function architecture applies equally to a commercial Transit van enrolled in Ford Pro Insure and to a personal F-150 whose owner might, someday, be offered an OEM-branded consumer policy. That is the reframing this grant supports: rather than reading it as defensive IP protecting a discontinued consumer program, it reads as infrastructure for the insurance business Ford is actively building through American Road, one where Ford's own subsidiary, not an outside carrier, decides how the telematics signal gets weighted before a rate is ever quoted. A carrier or MGA that wants to write usage-based policies on Ford vehicles going forward may find the aggregation layer itself, the part of the pricing chain an actuary would ordinarily control end to end, sitting inside Ford's patent claims rather than the carrier's own model.

Where the Grant Sits in the UBI Patent Landscape

Ford's claim is narrower and more technically specific than the OEM insurance patent application it was reportedly pursuing as early as 2022, which described a built-in system scoring drivers on ADAS feature activation, vehicle characteristics, and driving context to produce an aggregate insurance risk value for third-party data aggregators (Repairer Driven News, June 2022). That broader application does not appear among Ford's granted patents; the 2026 grant instead claims a specific optimization method, not the general concept of scoring a driver from vehicle data, a pattern consistent with the sharper Section 101 eligibility bar the USPTO has applied to software claims across the insurance sector this year. It also sits in a different layer of the UBI patent stack than the claims carriers themselves have been filing. State Farm's US Patent 11,257,146 covers incentivizing or penalizing vehicle renters based on telematics data collected during a rental or peer-to-peer rental period, a narrow, transaction-specific use of driving data rather than a general rating architecture. LexisNexis Risk Solutions, by contrast, had secured six telematics and connected-car patents by 2018 covering how driving data is ingested, normalized, stored, and distributed across what it said was 60% of global vehicle production at the time (LexisNexis Risk Solutions, August 2018), infrastructure patents for the pipe rather than the pricing math running through it.

Patent or Portfolio	Holder	Layer of the Stack	What It Actually Covers
US 12,694,456 (2026)	Ford Global Technologies	Rating math	Bayesian-optimized weighted aggregation of telematics signals into a UBI rate-prediction input, computed on the vehicle
US 11,257,146 (2022)	State Farm Mutual	Product/use case	Incentives and penalties applied to vehicle renters based on telematics data during a rental period
Six-patent portfolio (through 2018)	LexisNexis Risk Solutions	Data pipe	Ingestion, normalization, storage, and distribution of connected-car data across manufacturers
2022 application (unresolved)	Ford Global Technologies	Broad scoring concept	Driver/vehicle/context risk score from ADAS activation data for third-party aggregators; does not appear among Ford's granted patents

Whose Documentation Backs the Rate

Rate-filing regulation was not built for a pricing input a carrier licenses from a patent it does not own and cannot fully see inside. Texas already requires carriers to disclose specific information about any third-party model behind a filed rate: the vendor's name, the model's name and version, a description of the model, its inputs, and a description of how the model's output is used in ratemaking or underwriting, along with the rating variables that depend on that output (Texas Department of Insurance, PC441 exhibit). That disclosure regime was built around static, vendor-supplied catastrophe and predictive models. Ford's patented aggregation function is not static; claim 1's Bayesian loop can reoptimize the signal weighting against Ford's own archive between filing cycles, and neither a carrier licensing the output nor a state examiner reviewing the filing has visibility into the archive, the acquisition function, or the convergence criterion driving that reoptimization, because those details sit inside a patent claim describing the method, not inside a disclosure document a regulator can audit line by line.

The NAIC's ongoing telematics and big data work already flags the transparency gap that arises when a UBI score comes from a vendor a carrier does not fully control (NAIC Insurance Topics, Big Data). Ford's grant sharpens that gap rather than resolving it, because the vendor here is also the vehicle manufacturer with first-party sensor access no telematics app or plug-in dongle competitor can replicate, and because the method's own claim language builds in continuous, automated reoptimization as a design feature rather than an edge case. A carrier that licenses this aggregation function to support a UBI factor will need to reconstruct, for its actuarial memorandum, a description of a scoring methodology that changes on Ford's schedule, not the carrier's.

What happens next likely splits along the same fork Ford's own strategy has not yet resolved. If Ford licenses the aggregation-function IP to outside carriers and MGAs, it becomes functionally equivalent to what LexisNexis and Verisk already sell, a vendor score behind someone else's rate filing, just patent-protected and running on hardware only Ford controls. If Ford instead keeps the method inside American Road and its Ford Pro Insure fleet book, the rating math may never surface in an outside filing at all, reviewed only by whatever internal actuarial function Ford builds around its own captive-adjacent carrier. Either path puts a piece of the loss-cost signal that used to belong to the carrier's pricing team inside an automaker's patent portfolio instead.