Florida Sues First Street Over Climate Risk Scores and Home Values

Florida’s lawsuit raises a market question

Florida’s lawsuit against climate risk modeler First Street puts a market question in sharp relief: how much weight should homeowners, insurers and real estate platforms place on address-level hazard scores when the underlying estimates carry uncertainty? The answer matters because these ratings can turn technical projections into a simple number that appears easy to compare. Yet a clear score doesn’t necessarily mean the evidence behind it is equally clear at every property.

Filed Oct. 23, the complaint accuses First Street Technology Inc. of using flawed science to assign risk scores that could harm Florida homeowners. The company estimates exposure to flooding, wildfire, extreme heat, wind and other hazards, both today and decades ahead. MSCI Inc. recently acquired First Street and did not respond to a request for comment. The complaint’s claims remain allegations; the dispute will require scrutiny of the methods and marketing at issue, rather than treating either the state’s language or the company’s ratings as settled fact.

Scores reach buyers and insurers

First Street translates its estimates into scores from 1 to 10 and distributes them through real estate websites. Zillow, Redfin and Realtor.com began incorporating the ratings after the firm published its first major report in 2020. Insurers also use First Street data, according to Florida’s complaint, which argues that inaccurate scores could affect premiums or coverage. That chain of use matters: a model developed to describe hazards can influence decisions far beyond the people who build it.

That reach gives the dispute consequences beyond a technical debate. A score can shape how buyers assess a property and how sellers position it, while potential insurance effects matter in a state where households already face pressure from rising housing and insurance costs. A single rating may also compress several hazards and time horizons into one signal, leaving a reader to infer how the score was produced and what it means for a particular decision. The convenience is real. So is the risk that a simplified display will be read as a precise forecast.

Florida Attorney General James Uthmeier, a Republican seeking reelection, said on X that the data is inaccurate and that “homeowners deserve facts, not alarmist scores that hit their property values and their wallets.” The complaint describes First Street’s work as “climate chicanery” and accuses the company of “marketing fear to advance a climate activism agenda” while selling information about what it calls “fake” risks. That rhetoric makes the state’s position plain, but it doesn’t resolve the underlying question: whether particular methods or claims misstate what the data can support.

Accuracy at the property level is contested

The case follows growing scrutiny of how climate risk estimates are built, communicated and used. Modeling can identify broad patterns, but confidence tends to fall as analysis narrows to individual locations. Similar systems typically depend on assumptions about hazards, exposure and future conditions; choices at each step can affect the resulting estimate. A score can therefore be useful as a screening tool while still requiring explanation before it’s treated as a definitive judgment about one address.

In a 2024 comparison, Bloomberg News found that First Street’s 2020 flood estimates for Los Angeles overlapped with an academic team’s results only 25% of the time. That finding points to disagreement between two sets of estimates, but it doesn’t by itself establish why they differed or which one was more accurate. This is more significant than it looks: when different methods produce different readings, users need enough information to understand what a score measures, how uncertainty is handled and where the model’s limits begin.

Real estate platforms have already adjusted how they present the information. Zillow stopped displaying First Street scores directly on listings last year after objections from the California Regional Multiple Listing Service. Sellers and agents have also argued that the ratings could reduce property values and complicate sales. Presentation changes matter because the same data can carry a different impression when it appears as a prominent listing feature than when it’s presented with context about uncertainty and method.

Madison Condon, a Boston University law professor who studies climate risk, regulation and finance, has raised concerns about First Street’s marketing. She also warned that litigation could discourage the broader supply of physical climate risk data. Models involve uncertainty, she said, and disagreement among scientific models does not by itself show intentional deception. Here’s the thing: a market can need better disclosure and still depend on information that is imperfect. The hard part is distinguishing an honestly communicated estimate from a claim that overstates what the estimate can establish.

A second state challenges the modeler

Florida’s action follows a similar lawsuit Nebraska filed in August. Together, the cases add pressure on a data business whose ratings have become visible to consumers and relevant to insurance decisions. Connecticut’s insurance department, by contrast, partnered with First Street to provide residents free property assessments. The different approaches show that public institutions can treat the same kind of data as a possible consumer resource while other officials challenge its reliability and presentation.

The competing responses reflect the central tension: risk scores can make hazards easier to compare, yet their precision may suggest more certainty than property-level estimates can support. For insurers and real estate firms, the immediate issue is how to communicate model uncertainty when a score can influence a household’s financial choices. If you’re working in this space, the practical test is whether a consumer can tell what the rating covers, what assumptions shape it and how much weight to place on it. A number without that context may travel well, but it can leave the user with a false sense of clarity.

What the dispute could mean for users

The Florida case will test the claims made about First Street’s methods and marketing, while the broader debate concerns how climate data should be translated for everyday decisions. That said, the available facts don’t settle whether the disputed ratings are accurate at particular addresses or whether the company’s presentation crossed a line. The lawsuits may sharpen scrutiny of those questions, but uncertainty in a model is not, on its own, proof of misconduct.

For households, the immediate significance is more modest and more practical: a rating is one input, not a full account of a property’s risks or costs. For platforms and insurers, clarity about what a score can and cannot say is central to using it responsibly. And yet, if lawsuits make companies less willing to publish physical risk data, consumers could have fewer tools for comparing hazards. A careful distinction between uncertainty and deception will matter to both sides.