Insurance Curator WASHINGTON — Consumer groups, privacy advocates and some lawmakers warned this week that the rapid expansion of telematics-enabled auto insurance — known as usage‑based insurance or UBI — risks entrenching discrimination by turning drivers’ everyday movements into granular, monetizable profiles that can be resold and repurposed, even as insurers press the case that behaviour‑based pricing improves actuarial fairness and road safety. The debate has intensified amid government enforcement actions and a wave of state regulatory guidance in the United States, parallel privacy and uptake concerns in Europe and the United Kingdom, and fresh industry growth forecasts that industry officials say will make telematics an increasingly central pricing tool. (insurancenewsnet.com)
What happened: regulators and consumer advocates have ramped up scrutiny after reporting and enforcement revealed that connected‑vehicle and telematics data—including precise location, hard braking, acceleration and time‑of‑day driving—have been shared with data brokers and insurers in ways drivers did not expect. The Federal Trade Commission in January 2025 announced a proposed consent order against General Motors and its OnStar unit, alleging the company collected and sold drivers’ geolocation and driving behaviour data without adequate notice or affirmative consent; the FTC described the practice as “extremely privacy invasive.” (ftc.gov)
Why it matters: supporters of telematics say UBI lets insurers price on individual behaviour rather than proxies such as ZIP code or credit score, potentially reducing unfairness and rewarding safer drivers. Opponents — including Consumer Reports, the Consumer Federation of America and civil‑liberties groups — say the volume and persistence of the data collected, the opacity around algorithms, the resale of driver profiles by automakers and data brokers, and the use of opaque scorecards create new routes to disparate impacts on protected and vulnerable groups. Those groups say regulators must act now to prevent bias from becoming baked into pricing practices that increasingly rely on machine learning and third‑party data. (consumerreports.org)
Industry growth and the data trade
Telematics is no longer a boutique product: industry forecasts show adoption growing across Europe and North America, with some market analysts projecting tens of millions of telematics policies in the coming years. Research and Markets and other forecasters expect steady growth as smartphone‑based solutions and built‑in connected car systems proliferate. At the same time, consumer surveys and market research report persistent unease about privacy — a combination that insurers say complicates their commercial plans. (globenewswire.com)
The commercial mechanics are simple: insurers offer discounts or pay‑as‑you‑drive pricing in exchange for telematics data collected via an insurer app, an in‑vehicle plug‑in device, or directly from a carmaker’s telematics system. Data brokers and analytics firms such as LexisNexis and Verisk have bought, packaged and resold driving‑behaviour data or derived “driving behaviour history” files to insurers and other customers — a trade that U.S. senators and the FTC have scrutinized. In letters to regulators, senators said payments from data brokers to automakers amounted to pennies per vehicle, and raised concern that consumers were not given sufficiently clear notice. (repairerdrivennews.com)
Consumer harms and the discrimination risk
Consumer advocates say the core harm is not the existence of behavioural pricing but the combination of four factors: (1) the sensitivity and granularity of the data; (2) opaque proprietary scoring and pricing algorithms; (3) third‑party resale and long retention of driver records; and (4) the potential for seemingly neutral variables (time of day, route, frequency of hard braking) to act as proxies for race, income, occupation or other protected characteristics.
“Companies shouldn’t be using telematics data for marketing,” Justin Brookman, advocacy director for consumer privacy and technology at Consumer Reports, told investigators after reviewing insurer policies. Consumer Reports and the Consumer Federation flagged that companies vary widely in what they collect, how long they keep it and whether they reserve the right to repurpose it for claims, underwriting or marketing — practices that, advocates say, create downstream discrimination and privacy risks that consumers cannot easily control. (consumerreports.org)
The GM‑OnStar episode crystallised many of those risks for U.S. policymakers. The FTC’s complaint says data collected “as often as every three seconds” was shared with consumer reporting agencies and used in products that insurers could consult when setting rates. FTC Chair Lina Khan said the agency’s action was intended to “safeguard Americans’ privacy and protect people from unchecked surveillance.” The settlement would, among other things, ban GM and OnStar from disclosing covered driver data to consumer reporting agencies for five years and require affirmative consent before collecting certain connected‑vehicle data. (ftc.gov)
Proxy discrimination is not hypothetical. Academic and regulatory commentaries have warned for years that algorithmic risk models can reproduce and amplify historical injustices when predictors correlate with protected characteristics. An Oxford legal analysis and recent papers in the algorithmic fairness literature describe how opaque “big data” methods can mask disparate impacts and make it harder for watchdogs to detect unfairness. In insurance, those risks are sharpened because pricing is the primary product: small changes in a score can translate into meaningful premium differences for low‑income households or workers who must drive at night. (academic.oup.com)
Regulatory responses and the tug of war over fairness
Regulators are rushing to lay down expectations and guardrails — producing a patchwork of approaches that is reshaping insurer behaviour.
In the United States, the National Association of Insurance Commissioners in late 2023 adopted a model bulletin, “Use of Artificial Intelligence Systems by Insurers,” and many state insurance departments have issued or adopted similar guidance. As of March 2025, 24 states had adopted the NAIC model bulletin or similar rules requiring insurers to maintain written AI programs, test models for bias and document governance and vendor oversight. State officials say the model bulletin is meant to ensure accuracy, prevent unfair discrimination, and subject AI systems to traditional market conduct oversight. (skadden.com)
That regulatory momentum has provoked a countermovement in the industry. The National Association of Mutual Insurance Companies, representing many mid‑sized carriers, published an analysis calling AI‑focused legislation “unfounded” and warning that poorly designed restrictions could force insurers to abandon actuarially sound risk classification and worsen affordability and availability problems. “Insurance classifies based on risk, and insurance law requires those risk classifications to be actuarially sound and not unfairly discriminatory,” wrote Lindsey Klarkowski, NAMIC’s policy vice president for data science and AI. The industry trade groups say regulators must balance consumer fairness with the fundamental risk‑based purpose of insurance. (insurancejournal.com)
Across the Atlantic, European regulators and consumer bodies are similarly engaged. Supervisors and privacy authorities in the U.K. and EU have highlighted data protection as a primary barrier to consumer uptake, and industry surveys indicate a significant share of drivers who would otherwise consider UBI shy away over privacy and transparency concerns. In Britain, a 2025 consumer survey found more than a quarter of motorists citing privacy worries as a key barrier to telematics adoption. (lifeinsuranceinternational.com)
Insurers’ safety argument and empirical evidence
Insurance executives and telematics vendors stress UBI’s benefits for safety, targeted interventions and rewarding careful drivers. Cambridge Mobile Telematics, a leading provider of telematics analytics, has published multiple studies showing measurable reductions in risky behaviours among drivers who receive telematics feedback and incentives: their research shows highly engaged risky drivers reduced hard‑braking and distraction and would, the firm says, translate those behaviour changes into lower claim rates. The firm and several insurers argue that behaviour‑based pricing is intrinsically fairer than ZIP‑code or credit‑based surrogates. (cmtelematics.com)
But the empirical safety case does not erase distributional concerns. Consumer advocates point out that many telematics programs deliver discounts only to those who can comply with the digital requirements — smartphone ownership, stable connectivity, ability to keep a device powered and the time to engage with an app — effectively excluding people who are already economically marginalised. Moreover, variables such as “night‑time driving” penalise shift workers and essential workers who have no option but to commute when risks are higher, while routing data can penalise those who live in neighborhoods with poor road geometry or high traffic — conditions correlated with socio‑economic disadvantage. (insurancenewsnet.com)
Third‑party data brokers and the opacity problem
A recurring theme in recent investigations is what happens when automakers or telematics vendors sell or licence driving data to third parties. Multiple legislative inquiries and press accounts revealed that some automakers shared driving data with Verisk and LexisNexis, which then packaged the information into reports sold to insurers. Senators Ron Wyden and Ed Markey pressed the FTC to investigate, noting that data brokers bought driving records cheaply and could resell them without effective consumer control. The GM‑OnStar complaint and subsequent public debate helped demonstrate how connected‑vehicle data flows can escape the consent regimes consumers expect. (repairerdrivennews.com)
“The problematic practices we have uncovered and documented are likely just the tip of the iceberg,” the senators wrote in a July 2024 letter urging FTC enforcement. Industry defenders counter that data sharing can be limited and that properly governed telematics programs — with clear consent, limited retention and strict use controls — are valuable. But consumer groups say the burden is on companies and regulators to make those safeguards mandatory, not optional. (repairerdrivennews.com)
What consumer groups are calling for
Consumer advocates and civil‑liberties organisations have converged on a set of policy proposals they say would reduce the risk that telematics entrenches discrimination:
- Clear, affirmative consent and plain‑language disclosures about what data is collected, how long it will be kept, and whether it will be shared or sold beyond the primary insurer. (ftc.gov)
- Limits on data re‑sale to consumer reporting agencies or data brokers for insurance underwriting, or at minimum strict use constraints and accessible mechanisms for consumers to see and correct their data. (repairerdrivennews.com)
- Mandatory testing and independent audits for disparate impact and proxy discrimination in pricing algorithms and scorecards; regulators should require documentation and remediation plans. (skadden.com)
- Data‑minimisation and retention limits so that long histories of fine‑grained driving logs are not kept indefinitely and reused in ways consumers did not expect. (consumerreports.org)
Michael DeLong of the Consumer Federation of America, summarising his group’s white paper, said regulators must act before telematics becomes ubiquitous: “For too long there has been a lack of useful information about how the programs evaluate drivers and the scope of the data used,” he said. “Without strong oversight insurers could use telematics to unfairly discriminate against drivers based on qualities beyond their control.” (insurancenewsnet.com)
Insurers’ response and design choices
Many insurers acknowledge the reputational and regulatory risks and are adjusting product design and disclosure practices. Some carriers now offer clear opt‑in windows, shorter retention periods for raw telematics logs and promises to limit third‑party sharing. Others are pushing back more broadly, arguing that any restriction that weakens risk‑signal inputs — or forbids the use of variables correlated with risk — will raise rates for higher‑risk drivers and shrink coverage availability. “In setting rules of the road, policymakers must recognize that insurance is distinct in function and pricing from many other consumer products,” NAMIC wrote in its analysis. (namic.org)
Regulators, too, are wary of overreach. The NAIC guidance asks insurers to proportionately tailor governance to their use of AI and telematics and recommends model governance practices rather than prescriptive bans. States that have adopted the NAIC model bulletin say their goal is not to halt innovation but to create predictable expectations about governance, documentation and consumer notice. (skadden.com)
A narrow path forward — design, disclosure, and independent oversight
Experts interviewed for this investigation said a defensible regulatory path would combine three pillars: stronger privacy protections, mandatory disparate‑impact testing, and improved consumer control.
- Privacy: require affirmative, granular consent for connected‑vehicle features that could be resold; limit retention of granular location logs; prohibit sale to consumer reporting agencies without consumer opt‑in. The FTC’s proposed GM order enshrines some of these elements. (ftc.gov)
- Algorithmic testing: require insurers to run and publicly report independent tests for disparate impact, with regulators empowered to require remediation. The NAIC model bulletin and several state versions point in this direction. (skadden.com)
- Portability and transparency: give consumers a right to see, correct and delete their telematics profiles; require plain‑English explanations of how telematics scores affect premiums. Consumer groups say these are prerequisites for meaningful choice. (consumerreports.org)
Outlook: balancing personalization and fairness
Telematics and usage‑based pricing are likely to expand. Market forecasts point to steady uptake as more vehicles are sold with embedded connectivity and as smartphone‑based products mature. Technology can reward safer drivers and yield public safety benefits, as Cambridge Mobile Telematics’ studies suggest. But those benefits are not automatic and will be distributed unequally unless policymakers and industry designers confront the twin problems of opacity and downstream data trade. (globenewswire.com)
The stakes are practical and political. If telematics becomes the dominant pricing model without binding guardrails, critics warn, insurers may legalistically hide behind “behavioural” pricing while relying on opaque third‑party files and machine‑learned proxies that systematically disadvantage particular communities. If regulators overcorrect, industry representatives say, the result could be less precise pricing, elevated premiums for high‑risk groups, and reduced availability in marginal markets.
“Because, in the absence of proper controls, AI has the potential to increase the risk of inaccurate, arbitrary, capricious or unfairly discriminatory outcomes for consumers,” the NAIC bulletin warned, regulators must approach the issue “in a way that recognizes the distinct role of insurance while protecting consumers.” That balancing act — between personalization, privacy and the public interest — will shape whether telematics fulfills its promise to reward safer driving or becomes another vector for entrenched inequality. (skadden.com)
Methodology and sources
This article draws on regulatory filings and press releases, industry reports and forecasts, peer‑reviewed and white‑paper research from telematics providers, and investigative reporting by major news outlets. Key sources include the Federal Trade Commission’s proposed consent order against General Motors (Jan. 2025), NAIC publications and state adoption notices (through March 2025), Consumer Reports’ investigations and the Consumer Federation of America’s advocacy materials, Cambridge Mobile Telematics’ published studies on driver behaviour, and reporting and oversight letters from U.S. senators and other authorities. Where possible, statements were taken from public filings, press releases and published interviews. (ftc.gov)
— Reported from Washington and London.