After Montgomery v. Caribe, a broker's carrier-selection record is evidence. Here is what the decision says — and what a defensible record looks like.

A plain-English explainer of the Supreme Court's May 2026 decision on freight-broker liability for negligent carrier selection, a checklist for the record you should be able to produce, and a free check of any carrier's public FMCSA safety data.

What the Supreme Court decided

On May 14, 2026, in Montgomery v. Caribe Transport II, LLC (No. 24-1238), a unanimous Supreme Court held that a federal statute freight brokers had long relied on — the Federal Aviation Administration Authorization Act of 1994 (FAAAA) — does not shield brokers from state-law claims that they negligently selected an unsafe motor carrier.

The background

The FAAAA bars states from enforcing laws "related to a price, route, or service" of a motor carrier or broker. For years, brokers argued that a negligence suit over which carrier they hired was such a law, and that it was therefore preempted. Courts split. The FAAAA also contains a "safety exception," preserving states' "safety regulatory authority … with respect to motor vehicles," and the question was whether an ordinary state negligence claim about carrier selection falls inside that exception.

The holding

The Court held that it does. A state common-law negligent-selection claim is an exercise of the state's safety regulatory authority with respect to motor vehicles, so it survives preemption. Brokers can be sued in state court, under state negligence law, for hiring a carrier they should not have hired — and the case then turns on the facts: what the broker knew, what it could have known, and what it did with that information.

The nuance

The decision is about safety-based negligence. Several firms have noted that the reasoning does not reach every claim a shipper might bring: claims about cargo theft, double-brokering fraud or other purely economic losses do not fit the "safety … with respect to motor vehicles" exception in the same way, and remain subject to preemption arguments. Husch Blackwell's alert makes exactly this point. The practical line for a compliance program: the exposure the decision opens is about crashes and injuries, and the record that matters is the safety record.

What it does not do

Opinion: supremecourt.gov, No. 24-1238 (May 14, 2026).

Further reading (transportation-law alerts)

Not legal advice. This page is a general summary of a public court decision and public commentary, written by a data company, not a law firm. It is not legal advice and does not create an attorney-client relationship. Whether a given carrier-selection practice or record is reasonable in a given case is a question for your counsel under the law of the relevant state.

What a defensible carrier-selection record looks like

Every alert says the same thing: document your vetting. In a negligence case the question is not whether the carrier turned out to be unsafe — it is what the broker knew or reasonably could have known on the day it made the selection, and whether it acted reasonably on that. A useful record therefore has four properties.

1. Dated to the selection

The record shows what the public safety data said on the booking date, not what it says today. Data changes daily: authority is revoked, insurance lapses, out-of-service orders are issued and rescinded. A screenshot taken after the crash proves nothing about the decision.

2. Sourced and itemized

Each fact traces to a named public dataset (FMCSA census, inspections, crashes, out-of-service orders, operating authority, insurance filings) and each computed indicator shows its inputs. A bare number from a vendor with no breakdown is hard to defend and easy to attack.

3. Consistent

The same checks, applied the same way, to every carrier — with a written policy that says which flags stop a booking, which trigger review, and who reviews. Reasonableness is easier to show for a process than for a one-off judgment.

4. Retained and reproducible

Kept for at least the limitation period, in a form that can be produced in discovery, and reproducible from an archive if challenged. "We checked but did not save it" is the weakest position available.

Concretely, for each carrier and each selection you should be able to produce: the DOT number and legal name as verified; operating status and authority; insurance filing status; whether an out-of-service order was in force; the carrier's inspection and crash history relative to peers, with the indicators and their inputs; any red flags (for example a link to a prior, revoked identity); the date and time of the check; and the decision taken. CarrierScore's evidence report is one way to produce this record from archived public data; it is not the only way, and it does not replace your policy or your judgment.

Free check: one carrier, public FMCSA data

Enter a USDOT number. This runs a keyless, rate-limited call to the CarrierScore API and shows the carrier's two risk indices, their components, and any hard flags, computed from daily-archived FMCSA data. No account needed.

Free tier: 30 lookups per hour per IP. Higher = riskier on both indices (0–100, population percentiles).

Inspection / compliance risk

/ 100

    Crash risk

    / 100

      How the indices are built and how they were validated on held-out cohorts — including the numbers and the caveats — is published at carrierscore.io/methodology. Data: public FMCSA / U.S. DOT datasets. Not a safety fitness determination; not endorsed by FMCSA.