After an Uber or Lyft crash: whose insurance covers what
A rideshare crash runs on three coverage phases, not one. The phase at the moment of impact decides which policy answers first and who is likely responsible. This is the model, phase by phase, with the documentation habits that keep the claim provable. Specific company coverage amounts change and are not quoted here.
Rideshare coverage is built in three phases. Phase one is the app off, when the driver is a private motorist and only the personal auto policy is in the picture. Phase two is the app on with no ride accepted, when a limited contingent layer from the platform typically backs the personal policy. Phase three is a ride accepted or a passenger in the car, when a larger commercial layer from the platform applies. The phase at impact is the first question every claim will ask, the trip log answers it, and screenshots on your own device preserve the answer. Deadlines run on the state's clock either way.
Collision Bureau is not a law firm and this is not legal or medical advice. It is general information about what happens after a crash. For advice on your situation, talk to an attorney licensed in your state or a treating clinician.
A rideshare crash runs on three coverage phases
The thing that makes a rideshare crash different from any other crash is not the make of the car or the fact that an app opened the trip. It is that a rideshare driver can be in one of three insurance phases at any given moment, and the phase at the second of impact decides which policy is on the hook first. People who learn this after the crash spend weeks confused by letters from carriers that read like they contradict each other. People who learn it before the first call stop being confused, because the three-phase model is the same model every carrier and every platform works from, even when the headline numbers vary.
Here is the model in one paragraph, before the sections take each phase apart. Phase one is the app off: the driver is a private motorist, running errands or sitting at home, and the personal auto policy is the only thing in play. Phase two is the app on with no ride accepted: the driver is logged in and available but has no passenger and no match, and a limited contingent layer from the platform sits behind the personal policy. Phase three is the ride accepted or passenger on board: the driver is working on a specific trip, and a larger commercial layer from the platform applies. Three phases, two platforms that use this general shape, and one answer to the question of which coverage opens first.
Why does the industry split the trip into phases at all? Because the risk is not the same across the phases, and insurance has to price what it covers. A driver on the couch is not a commercial risk. A driver with a rider in the back seat is. The phases are the industry's way of saying so, and the practical result is that every rideshare claim starts with a stage question before any dollar question gets asked. The question is answered by a server, not by anyone's memory of the moment, because the platform logs the exact state the account was in when the crash hit.
Uber and Lyft are the two platforms Americans use most, and both operate under versions of the three-phase model as a matter of state regulation. Specific coverage amounts change over time, differ by state, and are a question for the policy in force on the day of the crash. This guide uses the names only to describe the model: whose rules the model follows, not what any single number is this quarter. If you need the actual limits on your day, the actual limits are a question for the platform's current public terms, your own carrier, and a licensed attorney reading the specific facts.
The reason the phase question matters before anything else is sequencing. A claim submitted to the wrong carrier in the wrong order wastes weeks, and the carrier that was supposed to answer first turns to the carrier that was going to answer second and asks why nobody told them. The driver, the passenger, and the third party outside the vehicle all have different versions of the same sequencing problem. The sections below take the sequence apart from each angle, starting with the phases themselves.
Rideshare coverage is a three-phase model. The phase at the second of impact decides which policy answers first. Every other question in the claim sits downstream of that one.
Phase one: the app is off
Phase one is the easiest to describe and the easiest to misread. When the rideshare app is closed, logged out, or set to off, the driver is a private motorist for the purposes of a crash. The platform is not on the hook, and the platform's commercial coverage is not in the picture at all. The personal auto policy answers, exactly the way it would if the driver had never installed the app in the first place. That part is clean. The misreading comes from drivers who assume that because they drive for a living, the platform is involved whenever they are in the vehicle. The platform is involved when the app says it is involved, and only then.
The practical consequence is simple and important. A driver in phase one carries exactly the coverage their personal policy carries, which is set by the declarations page, by their state's minimums, and by any endorsements they bought. If the collision was their fault and they carry only state-minimum liability, the limits are what the limits are. If someone else was at fault, the recovery comes from that driver's policy under the ordinary rules a non-rideshare driver would face. The identity of the driver as a rideshare driver has no bearing on what a phase-one claim does.
Where phase one gets interesting is the exclusion question. Many standard personal auto policies include a livery or driving-for-hire exclusion, which can disclaim coverage for a crash that happened while the car was being used commercially. In phase one the car was not being used commercially, so the exclusion does not apply by its own terms, but carriers sometimes probe the question before paying. The protection against that is the trip log. The platform's server knows the account was offline at the moment of the crash and can produce a record saying so. That record is one of the first documents a careful driver asks the platform to confirm after a phase-one incident, because the paperwork resolves the question faster than any argument on a phone call will.
Drivers in phase one who were on their way to turn the app on, or just finished a shift and turned it off, are still in phase one for the purposes of this question. The industry does not have a four-th phase called "on my way to work." The app is on or it is off. The server knows which, and the trip log settles it. Drivers who have a habit of leaving the app running during personal errands have, in effect, moved themselves into phase two for those errands, which is a different conversation covered in the next section.
If you are the driver and the crash happened in phase one, treat the incident like an ordinary personal auto crash from a reporting and documentation standpoint. Call your own insurer, make the report, and get the vehicle to a shop or lot through the normal path. The rideshare side comes up only if the account's status is later contested or if the platform asks about an incident that showed up in its own logs. Keep the proof of the app's state on the day, because the question of which phase applied is the one phase-one disputes always circle back to.
Phase one is the personal policy's problem, start to finish. Drivers who try to involve the platform in a phase-one crash can slow their own personal claim by weeks for no gain, because the platform has nothing to add to a crash the app did not touch.
Phase two: app on, waiting for a ride
Phase two begins the moment the driver taps "go online" or whatever the current button is called, and ends the moment a ride is accepted. In between, the driver is on the platform's clock in one sense and nobody's clock in another: available for a match, no match yet, driving around and burning gas while the app looks for a request. This is the phase that trips up the most people, because the driver is clearly "working" from their own perspective and the car looks like any other car to the outside world. The insurance picture during phase two sits in the gap between those two realities.
The structural fact about phase two is that both platforms carry a contingent liability layer that applies in this window. "Contingent" is the operative word. The platform's phase-two coverage usually sits behind the driver's personal policy, meaning the personal policy is asked first, and if the personal policy declines because of a driving-for-hire exclusion, the platform's contingent layer answers. The dollar amounts in phase two are typically lower than in phase three, and they vary by state because state regulators set the floor for rideshare coverage in each phase. The exact numbers are a question for the policy in force on the day of the crash and for an attorney licensed where it happened.
What matters in phase two is the ordering. A phase-two crash claim usually starts with the driver's personal carrier, exactly as it would in phase one. The personal carrier then asks the question the exclusion is designed to ask: was the driver carrying passengers for compensation at the time. The answer is no, because the match had not happened, but the answer is also not quite the same as phase one, because the app was logged in and available. Carriers evaluate the distinction with the platform's trip log in hand, and the paperwork settles it the same way it settled phase one: by consulting the server, not by taking anyone's word.
If the personal carrier declines because its policy excludes driving-for-hire including the app-on-waiting state, that is where the platform's contingent layer takes over. The driver files with the platform's insurer, often through a claims portal the platform runs, and the platform's carrier picks up the claim within the limits the state allows for phase two. The portal usually requires the police report, the driver's statement, and the trip log reference the platform itself produces. The timeline for a phase-two claim is longer than a comparable phase-one claim because the hand-off between carriers adds steps, which is a reason the documentation habits later in this guide matter more in phase two than anywhere else.
Third parties injured in a phase-two crash caused by the rideshare driver face the same ordering question from the opposite direction. The injured person's attorney evaluates which policy to put the demand to first, usually the driver's personal policy, with the platform's contingent layer named as additional available coverage if the personal policy declines. The injured person does not have to understand the model to file; the attorney does. Reading this section gives the injured person enough context to understand why early paperwork mentions two different insurers for one crash.
Phase two is the hand-off window. The personal policy is asked first, and the platform's contingent layer answers if the personal policy declines. Carriers consult the trip log to settle which phase applied, and the paperwork reflects both insurers.
Phase three: ride accepted or passenger on board
Phase three starts the moment the driver accepts a request and continues through the pickup, the trip, and the drop-off, ending when the ride is marked complete. During this entire window the platform's commercial coverage is the primary liability layer, in meaningful amounts, and the structural picture flips: the platform's insurer is on the hook first, and the personal policy sits behind it, often with no obligation to pay at all. This is the phase people instinctively picture when they hear "rideshare crash," and it is the phase with the clearest insurance answer.
The split inside phase three is sometimes called phase 2+ for the window between accepting a request and arriving at the pickup, and phase 3 for the window with the passenger in the car, but both operate under the commercial layer. The platform treats the acceptance as the start of the trip for coverage purposes because that is when the driver is contractually dispatched to a specific pickup. If a driver crashes on the way to a pickup and no passenger is in the car yet, the commercial layer still applies. The rider on the other end of the request has not entered the car, and that fact does not change the phase.
Commercial coverage in phase three is built from two layers, both specified by platform policy and state regulation. The first is liability, which answers to third parties, passengers, and other injured people when the rideshare driver is at fault. The second is uninsured and underinsured motorist coverage, which answers to people injured in the rideshare vehicle when another driver is at fault and that driver's insurance is missing or inadequate. The UM/UIM layer is why passengers in a phase-three crash still have a backstop even if the at-fault driver is uninsured and the rideshare driver has no fault. The exact structure and the exact limits sit in the platform's policy on the day of the crash.
Collision coverage for the vehicle itself in phase three is a different story. The platform may require the driver to carry collision and comprehensive on their own vehicle as a condition of being on the platform, and the platform's own coverage for the vehicle usually has a deductible that behaves like any other deductible. For the driver whose car is damaged in a phase-three crash, the practical path is often to file with the personal collision coverage first and let the carriers sort out reimbursement in the background through a process the industry calls subrogation. Not every driver has collision on the personal side; drivers who dropped it to save money and work in phase three are, in effect, working without physical-damage coverage on the days nobody is in the car.
From the passenger's perspective phase three is the cleanest of the three. The passenger did not cause the crash, so a liability claim runs against whichever driver is responsible, and the commercial layer answers for the rideshare driver when fault lands there. If the at-fault driver is on the outside of the trip and underinsured, the platform's UM/UIM layer steps in. The passenger does not file against the driver personally in any practical sense; the file runs against the platform's commercial coverage, which exists for exactly this scenario. The paperwork is extensive, and the next sections take it apart.
Who the responsible party is in each phase
Insurance phases decide which carrier opens the claim. The responsible party, which is the legal question of who is on the hook for the harm, is a separate question answered by fault. Phase one, phase two, and phase three all have the same fault inquiry underneath: who caused the crash. The phases do not change the answer to that question. What they change is who writes the check once the answer is known. People confuse the two often enough that it is worth saying slowly: the phase picks the policy; fault picks the person.
In phase one the responsible party inquiry is identical to any ordinary two-car crash. If the rideshare driver caused it, the rideshare driver's personal policy pays. If the other driver caused it, that driver's personal policy pays. If both share fault, states split the result under their own rules, and the split is a state-level question that varies more than any other piece of this picture. None of the phase mechanics touch any of that. The app being off makes the crash a private motorist crash in all respects.
In phase two the responsible party inquiry does not change either, but the carriers in the room do. If the rideshare driver caused it, the personal policy is asked first, and if it declines on the driving-for-hire exclusion, the platform's contingent layer answers within the limits the state sets. If the other driver caused it, the other driver's personal policy pays, and the rideshare driver's personal UM/UIM coverage can come into play when that other policy is thin or missing. Phase two also opens the question of whether the rideshare driver's own personal medical payments or personal injury protection coverage applies, because many of those coverages exclude active livery use and some do not. The answer is in the declarations page of the specific policy, which is why reading that page before anything else gets priced is a habit worth keeping.
In phase three the responsible party inquiry is the same, and the carriers in the room shift again. If the rideshare driver caused the crash, the platform's commercial liability layer answers for the third parties and passengers hurt, and the personal policy sits behind, often with no payment obligation at all. If the other driver caused the crash and is underinsured, the platform's UM/UIM layer is where the recovery for the passenger and the rideshare driver typically comes from. If the other driver caused the crash and is adequately insured, that driver's personal policy pays, and the platform's coverage may stay out of the money question entirely.
| Phase | If the rideshare driver is at fault | If the other driver is at fault |
|---|---|---|
| Phase 1 (app off) | Driver's personal policy | Other driver's personal policy; driver's UM/UIM as backstop |
| Phase 2 (app on, no ride) | Personal policy first; platform contingent layer if personal declines | Other driver's policy; driver's UM/UIM; platform UM/UIM where state law applies in phase 2 |
| Phase 3 (ride accepted or passenger) | Platform commercial liability | Other driver's policy; platform UM/UIM as backstop |
GENERAL MODEL. SPECIFIC AMOUNTS, UM/UIM OBLIGATIONS, AND ORDERING RULES ARE SET BY STATE LAW AND BY THE POLICY IN FORCE ON THE DAY OF THE CRASH.
The one place the phase genuinely changes the responsibility picture, as opposed to the policy picture, is in the question of who has deeper pockets to collect from. A plaintiff whose only option in phase one is a driver carrying state-minimum liability has a smaller recoverable pool than a plaintiff whose phase-three claim reaches the platform's commercial layer. The claim's merits are the same. The money that can be collected is not. That difference is a reason the phase question comes up early in every attorney review, because it decides what the claim is realistically worth without changing anything about who was at fault.
Driver, passenger, or third party: three different claims
A rideshare crash can produce injured people in three different roles, and each role runs its own version of the claim. The underlying model is the same, but the paperwork, the pressure, and the sequence feel different from each seat. Reading the next three paragraphs in the order of the role closest to your own gives you a map of the claim you are actually in, which is usually not the generic one described on the platform's support page.
The rideshare driver is the person with the most complex paperwork. The driver is a party to the crash, a party to a contract with the platform, and a party to a personal policy that may or may not want to be in the picture depending on the phase. If the driver is injured in a phase-three crash caused by another driver, the recovery runs against the at-fault driver's policy and, when that policy is thin, the platform's UM/UIM layer. If the driver is injured in a phase-three crash they themselves caused, their recovery options are limited because fault bars most paths, and the question becomes what their personal medical payments or injury protection coverage holds, if any. The platform's commercial liability layer is not there to pay the driver for a crash the driver caused.
The passenger is in the simplest position legally, and the one that strangers underestimate most. The passenger did not cause the crash in any version of events, so a liability claim runs against whichever driver is at fault, cleanly. If the rideshare driver is at fault, the file runs against the platform's commercial coverage. If the other driver is at fault, the file runs against that driver's personal policy, and if that policy is inadequate, the platform's UM/UIM layer answers next. From the passenger's seat the mechanics look like any injury claim: medical records, demand letter, negotiation. What is different is the volume of early paperwork, because the trip paperwork becomes evidence of what the passenger was doing in the car in the first place.
The third party is the person outside the rideshare transaction entirely: the driver of another vehicle hit by the rideshare car, the pedestrian or cyclist struck, or the passenger in a third vehicle involved. The third party's claim runs against whichever driver was at fault, with the phase mechanics deciding which policy answers for the rideshare side. For the third party the entire rideshare dimension can feel invisible until the demand letter identifies the platform's commercial carrier as the responsible payer. That identification matters, because commercial carriers negotiate differently from personal carriers, usually with more structure and more defense resources, and the plaintiff's early paperwork has to be built for a commercial adversary from the start.
| Role | Likely claim target | What makes it different |
|---|---|---|
| Rideshare driver | At-fault driver's policy or platform UM/UIM | Driver's own policy may exclude phase 2 or 3; income side runs on its own clock |
| Rideshare passenger | At-fault driver's policy; platform's commercial or UM/UIM layer | Fault almost never attaches to the passenger; trip paperwork is early evidence |
| Third party outside the trip | At-fault driver's policy; platform coverage if rideshare driver at fault | Commercial carrier involvement changes negotiation posture |
GENERAL PATTERNS BY ROLE. ACTUAL PATHS DEPEND ON PHASE, FAULT, STATE LAW, AND POLICY TERMS. A LICENSED ATTORNEY IN YOUR STATE READS THE SPECIFIC FACTS.
The gap between a crash and the first medical visit is the number insurers measure. Passengers who skip the first visit because they feel okay in the moment can find that gap read against the claim weeks later, when symptoms arrive and the record of the car ride ends at the drop-off. The trip receipt and a visit on day one together cost nothing and preserve both facts.
What the platform cooperates on, and what it does not
Platforms participate in post-crash paperwork in a specific, limited way. Knowing the shape of that participation prevents the two biggest misreads: expecting the platform to be a party and expecting it to disappear entirely. Neither is true. The platform is a defined participant with a defined role, and the role is bigger than people think in some ways and smaller in others.
What the platform cooperates on, consistently: confirming the account status at the moment of the crash, producing a trip stage identifier when a trip was in progress, routing claims to its commercial insurer when the phase applies, and providing access to in-app support threads and ride receipts when the account holder requests them. These are the records the platform owns and the actions the regulations require. Carriers know how to request them, and attorneys who handle rideshare claims regularly know the request paths and the response times. Access tends to be fine when the proper request is made from the proper party in writing.
What the platform does not do, equally consistently: provide information about the other driver outside the trip, provide passenger contact information to the other driver's insurer, make substantive statements about fault, or stand in as a party to the lawsuit when its driver is sued. The platform's position is that its driver is an independent party, and the platform's role is to make the trip data and the commercial coverage available. That position has been litigated at length in multiple states; the outcomes vary, and the practical reality on an ordinary claim is that the platform's cooperation runs through its insurer rather than through its support channel. A support chat is a bad place to settle a claim question. A written claim to the commercial carrier is the correct place.
The piece people miss most often is retention. Platforms do not keep ride data forever. The exact windows change over time and vary by data type. Some data is retained for a short period after a trip and some for a longer period, and the platforms publish retention policies that are the actual rules on any given day. The practical consequence for an injured person is that the earlier a formal preservation request goes in, the better the odds that the specific records in question are still available when the file is ready to use them. The free attorney call is the fastest way to send that request through the right channel, because the written preservation letter from a law office carries different weight than a user tapping "contact support" in an app.
Platforms also run their own investigations after serious crashes, sometimes called trust and safety reviews. These reviews can result in account pauses, deactivations, or reinstatements. They are not legal proceedings. They do not decide fault. They do decide whether the driver can earn income on the platform while the file is open, which is a separate concern that intersects with the claim in the lost-income category but does not change the claim's structure. Drivers whose accounts are paused during a review should document the dates and the correspondence; passengers do not need to track those details, because they do not touch the passenger's case.
The ride data the app logs on its own
Here is the quiet advantage every rideshare crash has over an ordinary two-car crash: the platform's server already recorded most of what the aftermath needs to prove. The exact dataset varies by platform, but the categories are stable. Account status at a given timestamp, meaning whether the driver was in phase one, two, or three. The trip identifier if a trip was active. Pickup and dropoff coordinates and times. Route taken, as a sequence of location pings. Payment status. In-app messages between the driver and the rider. In some cases, driver phone sensor data captured through the platform's safety features. The server knows all of it, and none of it depends on anyone's memory of what happened.
The reason this matters is that disputes about rideshare crashes often turn on timing, in two different ways. The first is phase timing: did the crash happen before the driver tapped "accept" on the request, or after. The server knows, and the answer is not a question of credibility. The second is route timing: at a given second, where was the car and how fast was it moving. The ping sequence is a decent, though not perfect, record of both, and when combined with any dashcam footage or other sources, it typically settles the position question better than any witness statement would.
What the server does not log, because servers do not have eyes, is the fault question. The server knows the car was at a given intersection at a given moment moving at a given approximate speed. It does not know whether the light was green or red. It does not know whether the other driver turned without yielding. It does not know what the rideshare driver did or did not do in the three seconds before impact. All of that is still a human-evidence question, built from the police report, witness statements, photos, and footage. The server data complements those sources without replacing them.
Access to the server data is a request process, not a self-service. The driver can see some of it through the driver app's history. The passenger can see less through the rider app, mostly ride receipts and the route. A claim or an attorney can request more through the platform's claims channel, usually accompanied by a signed release from the party on the trip. The requests are routine and the platforms handle them regularly, but the response times are in weeks, not days. That lag is one more reason the preservation request should go out early rather than late.
The piece of app data that moves fastest, and that people forget about, is in-app messages between the driver and the rider. Some messages are kept for a short window after the trip and then purged. If a message in the hours before the crash is relevant, screenshotting it on your own device is the move that preserves it independently of the platform. The screenshot sits on your phone and in your cloud account for as long as you keep it. The server's copy sits on the server for as long as the platform's retention policy holds it, which is not forever.
The platform's server already knows the phase, the trip ID, the route, and the pings. Those records settle the questions human memory cannot settle. They do not settle fault. And they do not stay available forever, which is why preservation requests go out early.
Screenshots and receipts in the first week
The documents that survive a rideshare crash are the ones you captured on your own device while the information was fresh. The platform's own copies exist on servers you do not control, under retention windows you did not set, and behind access processes that can take weeks. Your copies are in your pocket within seconds of a tap. This is not a redundant habit; it is the difference between a case built on your own paperwork and a case built on a records request that may or may not come back complete.
For the passenger, the first week's screenshot list is short and specific. The ride receipt that lands in the app after the trip ends. The trip detail screen, including the map of the route, the pickup time, the drop-off time, and the fare. The driver profile, including the driver's first name and the vehicle description. The in-app support thread, if you opened one. Any messages exchanged with the driver through the app. Each of these takes a second to capture, and taken together they establish the trip, the phase, and the parties, which are the three facts a claim needs on page one.
For the driver, the first week's list is longer because the driver has more operational data. The acceptance screen for the trip, the dispatch time, the earnings line for the trip, the trip stage history if the app exposes it, the driver support thread, and the vehicle inspection history if the platform keeps one. Drivers also screenshot the app's "online" dashboard around the time of the crash, because the dashboard sometimes shows status transitions that are useful later. Dashcam footage, when it exists, is downloaded off the camera's SD card to a device you own, not left on the camera to be overwritten by the next loop.
For the third party, the list is harder because the third party does not have access to the trip. The move is to document the vehicle from the outside: a photo of the rideshare sticker or trade dress on the car, the make and license plate, the driver's name as it appeared on any paperwork exchanged at the scene, and the trip receipt or trip identifier if the driver or passenger shares it. If none of that is available, the police report becomes the main source for identifying which platform the car was on, and the attorney's follow-up requests fill the rest in.
| Role | What to screenshot or save | When |
|---|---|---|
| Passenger | Ride receipt, trip map, driver profile, in-app messages, support thread | Within hours of the crash |
| Driver | Acceptance screen, earnings line, trip stage history, support thread, dashcam SD card | Same shift, before the next trip overwrites screens |
| Third party | Rideshare sticker, plate, driver's name from exchange, trip ID if shared | At the scene or in the first hour |
THESE ARE DOCUMENTATION HABITS, NOT A COMPLETE EVIDENCE LIST. STATE RULES AND INDIVIDUAL FACTS DECIDE WHAT ELSE MATTERS.
Store the screenshots in two places. The device is one place, and devices break, run out of storage, and get replaced. The second place is a cloud account you control, under an email you check. Email the screenshots to yourself if you have no other cloud. The point is that the evidence has to survive a dropped phone, a stolen phone, or a phone that gets factory-reset by someone trying to help. One backup is a habit; two is insurance against the week going worse than you expect, which is a thing that happens more than people plan for.
The first 72 hours, in order
Here is the sequence a careful first 3 days follow after a rideshare crash. Treat the day counts as an editable example of a common pattern, not a mandatory schedule, because your situation sets the real sequence and nothing on a page can predict the specific calls that come in.
| Window | Action | Why it matters |
|---|---|---|
| Hour 0 to 1 | Scene documentation, 911 if anyone is hurt, screenshots of the app | Evidence is freshest right now; the server state matters, and so do witnesses |
| Hour 1 to 6 | Medical visit if anyone is hurt; crash report filed | The gap between crash and first visit gets measured by insurers; the report starts the paper trail |
| Day 1 | Notify your own insurer; passenger reports through the app; driver reports through the driver app | Both carriers and the platform need the notice; the earlier it goes in, the cleaner the file |
| Day 1 to 2 | Make a one-sentence note of what happened and keep it; save receipts | The written note on day one reads better in month six than a reconstruction will |
| Day 2 to 3 | Free attorney call; attorney sends preservation letter to the platform | Trip-stage data and in-app messages have shorter retention windows than people expect |
EXAMPLE SEQUENCE ONLY. YOUR OWN SITUATION, INJURIES, AND LOCAL RULES SET THE REAL CADENCE.
The point of the sequence is not that every box has to be checked on time. The point is that each item on the list exists for a reason tied to the model: evidence that decays, deadlines that start, and carriers that work faster when the paper comes in clean. Missing day one does not kill a claim. The claim just has to be rebuilt from fewer early documents, and the office doing the rebuild will quietly spend hours on reconstruction that early habits would have avoided.
What your own policy says about rideshare driving
Rideshare drivers get asked the same question by every attorney at the first call: what does your personal policy actually say about driving for a platform. The question lands cold because most drivers have never read their own declarations page and never read the exclusions attached to it. The answer is in both documents, and both are already in your email or on your carrier's app. Reading them takes 15 minutes. The reading saves months of surprise on the back end of a claim.
Standard personal auto policies in most states include some version of a livery, driving-for-hire, or public conveyance exclusion. The language is not identical across carriers and the application is not identical across states, but the structure is: the policy covers private use of the vehicle, and if the vehicle is being used to transport passengers for compensation, coverage can be limited or denied. Rideshare driving falls inside that definition under most policies, which is why the question matters. Some carriers extend coverage into phase two but not phase three. Some exclude both. Some sell a rideshare endorsement as a rider that fills part or all of the gap, at an added premium.
The rideshare endorsement is worth asking about by name, because carriers do not always surface it. The endorsement does not usually turn a personal policy into full commercial coverage; what it does is close the exclusion for the phases the endorsement covers, so a claim in those phases does not get denied on the exclusion. It is cheaper than commercial auto insurance and more expensive than the base personal policy, and whether it is right for a given driver is a conversation with the carrier that costs a phone call and no money.
Drivers who learn after a crash that their policy excluded the phase they were in have a hard month. The personal policy declines, the platform's contingent layer handles its share, and anything above that layer is a gap the driver carries personally. There is no magic to this part. The exclusion is in the contract, the contract was signed, and the only lever available after the fact is whether the exclusion applies cleanly to the specific crash. That is a question for a licensed attorney reading the actual policy language, not a question for anyone's assumption about what carriers "usually" do.
For passengers and third parties the question looks different, because the exclusions on the rideshare driver's personal policy do not usually block their recovery. The platform's commercial layer during phase three is designed to answer in exactly the situation where the personal policy would try to walk away. The injured passenger or third party does not need to read the driver's policy; the carriers and attorneys sort out the policy stack in the background. What matters at the role level is that the money has somewhere to come from, and in phase three the money has a defined source that does not depend on the driver's exclusion drama.
The phase question has an answer. Ask it before the paperwork hardens.
One request covers the attorney, the tow, the repair, and the rental. It costs you nothing, ever.
Three callers, three different calls
The structural pieces above are the same for everyone involved in a rideshare crash. The way those pieces land is not, and the fastest way to see the range is to walk three callers through the same free review door. All three callers below are composites invented here for illustration and are not real people or real cases. Each shows how the same model reads three different starting positions.
Caller one was a rideshare driver in phase three when another driver ran a red light and hit the car at speed. There is a police report with the other driver cited, a passenger in the back seat who is also now injured, and clean dashcam footage of the light. This call runs fast on the structural questions. The phase settles the carrier order: the platform's UM/UIM layer sits on top for the driver's and passenger's injuries if the other driver is underinsured, and the other driver's liability policy answers first if it has limits. The dashcam and the report settle the fault question before anyone argues. The driver's call becomes a records-and-patience case, with a separate conversation about the account pause the platform's safety review may trigger.
Caller two was a passenger in a phase-three car when the rideshare driver got into a low-speed intersection crash with another vehicle, both drivers blame each other, nobody called the police, and the passenger's neck started hurting the next day. This call runs slow on fault. The driver and the other driver have conflicting stories, no citation was issued, and the server data confirms the car's position without resolving who had the right of way. The call also runs through the trip paperwork quickly: the ride receipt, the driver profile, the in-app support thread, and the phase confirmation move fast because the passenger controls them. The attorney conversation becomes a question of whether a camera canvass of the intersection will produce footage before the loop overwrites, which is a case where the first few days of the aftermath decide whether the claim exists in a provable form.
Caller three was in another car hit by a rideshare driver during phase two: the driver was logged in but had no passenger. The caller's bumper and knee are both bruised, and the rideshare driver's personal policy sent a reservation-of-rights letter citing the livery exclusion. This call runs through the sequencing: the personal policy's position is a signal that the platform's contingent layer may be where the recovery comes from, and the attorney's job is to put the claim in the proper order so neither carrier can point to the other indefinitely. The paperwork volume is larger here than in a comparable non-rideshare crash because two insurers are evaluating the same facts under different theories at once.
Three callers, three completely different calls, one model underneath. Each caller left the review knowing which carrier opens the claim, which paperwork their own copies cover, and which deadlines their state imposes. None of that cost anything to learn, which is the useful part. All three also learned the same practical thing, which is that the free call changes nothing by happening and preserves the ability to answer every one of the questions it asks.
Medical bills in a rideshare crash
A rideshare crash raises the same medical-bill question any crash does, with one extra layer that trips people up. Here is the clean version. Health insurance usually acts as first-pass payer while the claim is open, meaning the clinic bills your health plan, your plan pays its share, and the balance is tracked for later. In states that have personal injury protection or medical payments coverage, those layers can also come in early. The liability recovery at the end, from whichever carrier owes it, is what resolves the final numbers under the health plan's lien or subrogation right.
The rideshare layer that confuses people is the question of whose medical payments coverage is in play. If you were a passenger, the platform's commercial coverage during phase three can include a medical payments component that pays some bills early regardless of fault. If you were the driver, your personal medical payments or personal injury protection coverage may or may not apply depending on whether it excludes active rideshare use. If you were the third party, your own personal medical payments coverage is in the picture, and whichever liability carrier opens the file eventually pays its share through the ordinary mechanisms.
None of this changes the practical move, which is to get seen because you are hurt. The medical question is a clinical question answered by a clinician, not an insurance question answered by a carrier. A clinic needs to know you were in a crash; it does not need you to have the carrier picture figured out. Give the clinic the information it asks for, keep the records that come out of the visit, and let the carrier mechanics work in the background. The one habit that matters is not performing toughness in the room. The record of what you tell the clinician becomes the medical history the claim gets read against, and understating what hurts in the moment can quietly cost later.
The second habit is the one every medical pillar on this site repeats, because it matters most here too. Do not wait days to be seen because the paperwork feels complicated. The gap between the crash and the first visit is one of the numbers insurers measure most. A long gap gets read as evidence the injury was light, regardless of what the symptoms were doing in those days. A visit on day one, even if the visit is brief, closes that gap and starts a dated record that the claim will later be built from.
The lien side of medical bills in a rideshare case can be larger than in a non-rideshare case, because commercial coverage limits are sometimes high enough to attract more aggressive subrogation posture from health plans. That is a question your attorney at the end of the file knows how to negotiate, and the paperwork trail that supports the negotiation is the exact paperwork the medical visits generated. In other words, the first visit matters twice: once for the injury record and once for the lien math at the end.
The deadline logic still applies
The commercial dimension of a rideshare claim can make it feel like the ordinary rules of personal injury law got replaced. They did not. The deadlines, the fault systems, and the comparative rules all run on exactly the same clocks as any other crash in the state where the crash happened. Rideshare does not pause anything. The commercial carrier does not get a longer window than the driver's personal carrier would. The state's deadline is the state's deadline.
Every state sets a statute of limitations on injury claims. Two years from the crash is common, and some states allow less. The statute runs from the date of the crash in most situations, and it runs regardless of which carrier is on the hook, how many carriers are involved, or how long the trust-and-safety review takes on the platform's side. The free review tells you what your state's number is and how much runway remains in your specific case.
Rideshare adds a second clock that is not a legal deadline but behaves like one. Platform-held data has retention windows that are measured in weeks and months rather than years. Trip-stage logs, driver app history, in-app messages, and dashcam footage all have their own windows. A preservation request made at month twenty can arrive after the window closed on the specific thirty-second clip that would have settled the question. A preservation request made in week one can lock in exactly that clip for the whole life of the claim. The statutory deadline is the last day to file. The retention deadline is the earlier day the evidence has to be caught before it goes.
There is also a notice window that varies by state and by situation. Some states impose shorter notice requirements for claims involving commercial carriers, government vehicles, or specific industries, and rideshare has drawn its own regulatory frameworks state by state. The exact rules change, which is why state-specific numbers do not belong on a national page; it is also why a licensed attorney in your state reads the specific rules for the specific day of the crash. The clean move is the same move the whole guide has pointed to: ask early, while the paperwork can still carry the facts the law will later ask about.
None of this is urgent in the way sales pitches try to make it urgent. None of this is a reason to pick an attorney by billboard or to sign a release the first week to make the mess stop. All of it is a reason to make the free call before much more time passes, because the free call does not commit you to anything and because the free call sends the first preservation requests the file eventually depends on. A week does not break any claim. A missed year can.
It does not. The statute runs the same way whether the responsible carrier is a personal auto carrier or a national commercial program. The retention windows for trip-stage data and in-app messages are shorter than people expect, so the practical deadline for preserving the evidence is often weeks, not years.
Sequencing mistakes that compound the problem
The costliest mistakes in a rideshare aftermath are sequencing mistakes: right actions taken in the wrong order. The three-phase model, the retention windows, and the multi-carrier paperwork all compound errors that would be small in an ordinary crash. The point of a sequence is not that there is one correct order for everyone; it is that some orders cost you more than others, and knowing which ones does not require any technical knowledge.
The first sequencing mistake is giving a recorded statement to a carrier that is not your own before anything else happens. The other driver's carrier calls fast, sometimes within hours, and the question is almost always framed as helpful. The statement becomes evidence in a case the person giving it did not know had started. In a rideshare crash this gets worse, because the commercial carrier for one side and the personal carrier for the other may both be asking in the same week, and the person on the phone cannot easily tell which call is which. The clean sequence is to make the free attorney call first and bring exactly this question to it, with your state and your role on the table.
The second sequencing mistake is handling the vehicle before the documentation is complete. Rideshare drivers whose vehicles are their livelihood are tempted to get the car back on the road quickly, which is understandable and sometimes unavoidable. What it costs, when the paperwork has not caught up, is physical evidence. A car that goes to repair before anyone photographs it from six angles loses whatever it carried about the fault question. Dashcam SD cards that stay in the camera get overwritten by the next shift. Both are preventable with a half hour of habit, and both are expensive in minutes to redo later.
The third sequencing mistake is signing early paperwork from any carrier without reading it carefully. Early releases, broad medical authorizations, and settlement offers in the first days of a claim all exist, and each has a specific later cost. A broad medical authorization can reach records from years before the crash, and those records become arguments about preexisting conditions. An early settlement is a release, which generally ends the claim at whatever number it named, before the medical picture has had a chance to develop. Both are one-way doors. A free attorney call before signing is the way to see where the door leads.
The fourth sequencing mistake, specific to rideshare, is treating the platform's support channel as the right place to resolve a claim question. Support is for operational issues: a missing receipt, an account question, a trip dispute. It is not where claim-value conversations happen. The platform's commercial carrier runs those conversations through its own claims channel, and attorneys who handle rideshare claims know the proper paths. Hours spent in a support chat about the crash are hours that produced nothing in the file that actually decides what gets paid.
Rideshare vehicles are often back on the road fast because the vehicle is the income. The cost of that speed is physical evidence that walks away before anyone preserves it. 30 minutes of photos from six angles, saved to a cloud account you control, is the cheapest insurance there is against a fault argument you did not know was coming.
Where the conversation starts
Everything above sorts into a small list of actions in a specific order. Preserve the paperwork on your own device. Get seen if you are hurt. File the notices on the carrier and platform side that your role requires. Make the free attorney call, which sends a preservation letter to the platform before the retention windows run out. Read everything that arrives in your mail. Keep every receipt and every record, in a folder that costs nothing and quietly becomes the claim. That is the whole list.
The first conversation is the one almost everyone benefits from, and it is also the cheapest step in the sequence. A free attorney call does nothing by happening. It commits you to nothing. It costs you nothing. It answers the four questions you did not know to ask: which phase the server has you in, which carrier opens the claim, how much time your state leaves, and what the first preservation letter should capture. Making the call in week one is not the same as making it in month six, because week one has the evidence that month six does not.
Rideshare crashes have a reputation for being complicated. They are not more complicated than other crashes; they are more layered. The layers are the three phases, the two carriers that can be in the picture at once, and the server data that both helps and expires. Each layer has a clean rule you can see from the outside once the model is in your head. The purpose of this guide was to put the model in your head early, so the aftermath is a sequence rather than a surprise.
Collision Bureau exists because people in the middle of this week need one phone call to cover the lanes that run on different clocks: the attorney, the tow, the repair, and the rental. The attorneys we route to review the facts under their state's rules at no cost. The call is where the model becomes a specific answer about your crash, which is the only version of this that matters.
Questions people actually ask
01Which insurance answers first in a rideshare crash?
It depends on the phase the app was in at the moment of impact. With the app off, the driver's personal auto policy answers first and the platform's coverage is not in the picture. With the app on and no ride accepted, a limited contingent layer from the platform may apply when the driver's personal policy declines. With a ride accepted or a passenger in the car, a larger commercial layer from the platform applies. The phase is the first question any claim will ask, and the trip log answers it.
02I was a passenger in an Uber or Lyft when we were hit. Whose insurance pays?
Passengers are in phase three by definition, so the platform's commercial coverage is in the picture for the ride itself. Which carrier actually pays, the rideshare driver's side or the other driver's, depends on fault. As a passenger you did not cause the crash, so a liability claim runs against whichever driver was responsible. If the other driver was at fault and underinsured, the rideshare platform's uninsured and underinsured motorist coverage can be the backstop. Keep the ride receipt and the trip ID.
03Does my personal auto policy cover me while I drive for a rideshare app?
Many standard personal auto policies exclude driving for hire, which includes rideshare, and some add it back only through a rideshare endorsement sold as a rider. Read your declarations page and the exclusions section, or call your carrier and ask the question plainly. The answer varies by carrier and state. The practical result is that drivers who skip the question often learn it the hard way after a claim, when a denial letter cites an exclusion they never knew they had agreed to.
04What should I save from the app after a crash?
Screenshot the trip screen, the receipt, the map of the route, the driver profile if you were a passenger, and any in-app support thread. Save them to your own device and to a cloud account you control. The platform logs the same data on its side, but access to that log depends on a request process you may not control and a retention window you did not set. Your own copies are the ones that are there in month six when the claim is still open.
05Do Uber and Lyft give out the other driver's insurance information?
They give the parties on the trip the information their own systems hold, which is the rideshare driver's identity and the platform's commercial coverage for the trip. They do not hold the other driver's personal insurance information. That comes from the exchange at the scene, from the police report, or from a records request later. If nobody exchanged information at the scene, the police report and witness statements become the main sources, which is a reason the report matters.
06I was the rideshare driver and the passenger is hurt. What happens to me?
The platform's commercial coverage applies to the trip, which includes injuries to the passenger when the driver's fault is in the picture. Your own exposure depends on the phase, the policy, and the facts, and it is exactly the question to bring to a licensed attorney in your state before giving a recorded statement to anyone. The platform may deactivate your account during an investigation. That is a labor and contract question, not a legal one, and it is separate from the claim.
07What if the other driver caused the crash and has no insurance?
The platform's commercial coverage during phases two and three typically includes uninsured and underinsured motorist coverage, which exists for exactly this scenario. The amount of that coverage varies by phase and state. Your own uninsured motorist coverage on your personal policy can also come into play depending on who you are and what you were doing. The order the layers answer in is a question for a licensed attorney, who reads the actual policy language, not a generic summary.
08Can the platform deactivate the driver because of the crash?
Platforms can and sometimes do pause an account during an investigation. The practical effect is that the income side of the aftermath becomes its own problem, separate from the injury claim and the vehicle claim. The claim moves on its own clocks regardless. If the account is paused for a stretch, document the dates, because lost rideshare income is still lost income that attaches to the claim when fault supports it, and a licensed attorney in your state can answer whether and how.
09How is a rideshare passenger's medical bill handled while the claim is open?
Health insurance usually acts as first-pass payer, and the health plan is paid back from any recovery under a lien or subrogation right. In states that have personal injury protection or medical payments coverage, those can also come in early. The platform's coverage does not pay bills in real time; liability claims settle at the end. Treat records as the single most important paperwork, because the records decide value and the first-pass payer protects access to care.
10Is there a deadline on a rideshare injury claim?
Yes. Every state limits how long you have to file an injury claim. Two years from the crash is common, and some states allow less. Claims against commercial carriers do not pause the deadline. Evidence tied to a specific trip can be short-lived: trip-stage logs, driver app history, dashcam video, and ride receipts are not held forever. The free attorney call tells you how much time your state leaves you and what to preserve first.
You now know the model. Make the call while the evidence is fresh.
One request covers the attorney, the tow, the repair, and the rental. It costs you nothing, ever.