Georgia Supreme Court Says Cops Need Warrants To Search Vehicle Crash Data Recorders

from the enhanced-privacy-expectations dept

A couple of years ago, the Georgia state appeals court interpreted the Supreme Court’s Riley decision to cover data pulled from vehicles after accidents. If warrants were required to search cellphones — thanks to their ability to store all sorts of personal information — it stood to reason warrants should be needed to access other data not visible to the naked eye.

Extracting and interpreting the information from a car’s black box is not like putting a car on a lift and examining the brakes or tires. Because the recorded data is not exposed to the public, and because the stored data is so difficult to extract and interpret, we hold there is a reasonable expectation of privacy in that information, protected by the Fourth Amendment, which required law enforcement in the absence of exigent circumstances to obtain a warrant before extracting the information from an impounded vehicle.

The court told law enforcement that warrants should be considered must-haves in nearly every situation involving data belonging to someone else. The contents of a vehicle’s black box are tied to the vehicle, not any third party. In addition, the data recording is mandated by the government, making this an involuntary process. The Third Party Doctrine only covers information voluntarily given to third parties. The data collection that occurs when a vehicle’s air bag deploys isn’t voluntary.

Georgia’s highest court has now come to the same conclusion, but in a different case. (via It doesn’t cite the Riley decision, but it does point to the more recent Carpenter decision, which established a warrant requirement for historical cell site location info. Not all black boxes record GPS location, but that’s not really the point. The point is the information belongs to the car’s owner. In order for the government to retrieve and interpret it, it has to enter a person’s vehicle, attach a device to the car, and download the info. All of these steps implicate the Fourth Amendment, even though this particular issue hasn’t been litigated at the federal level.

The key here is the expectation of privacy. Also important in this case is the testimony of officers who undercut their own claims of inevitable discovery by asserting they almost never sought warrants to obtain black box data. A warrant was obtained one day after the data had already been downloaded and examined. The state Supreme Court doesn’t consider applying for a warrant after the fact to be the same thing as applying for a warrant before performing a search.

The decision [PDF] follows the steps of the investigation, noting the events that undercut the lower court’s application of the inevitable discovery warrant exception. This is what happened at the crash scene:

Investigator Jason Hatcher entered the passenger compartments of both vehicles, attached a crash data retrieval (CDR) device to data ports in the cars, and used the CDR to download data from the ACMs. The data retrieved from the Charger indicated that, moments before the collision, Mobley was driving nearly 100 miles per hour.

The warrant wasn’t actually obtained until the next day when a different investigator joined the investigation.

The next day, Investigator Bryan Thornton joined the team of officers investigating the collision. He discussed the case with the officers who had responded to the crash, visited and personally inspected the scene of the collision, and then applied for a warrant to search the Charger and Corvette and to physically remove and seize the ACMs from both vehicles. When Investigator Thornton made his application for a warrant, he was aware that Investigator Hatcher already had retrieved the data from the ACMs and that the data indicated that the Charger had been traveling at an excessive rate of speed.

The court notes that the warrant application did not rely on data already obtained by investigators to establish probable cause for the search. But it also notes Inspector Thornton added something to his warrant application that could only have been discovered via the search that had already occurred: the Charger’s high rate of speed.

[A]lthough he did not testify explicitly about his knowledge of the substance of that data at the time he applied for a warrant, he said that the data was the basis for the conclusion that Mobley committed vehicular homicide in the first degree (as opposed to some lesser offense or no offense at all). In the warrant application, Investigator Thornton represented that the ACMs likely contained evidence of vehicular homicide in the first degree. His testimony at the hearing implies, therefore, that he had a meaningful awareness of the substance of the data when he made the warrant application.

The officers argued recovery of crash data was routine in these investigations. This is likely true. But it’s also true investigators almost never obtained warrants for this data, making it far less likely the data would have been obtained Constitutionally. A certain amount of intrusion into a person’s property — in this case, a car — is needed to retrieve the data, which should have put officers on notice warrants were necessary.

Although Mobley disputes the idea that he had no reasonable expectation of privacy in the data retrieved from the ACM on the Charger, we find it unnecessary to resolve that question. To retrieve the data, Investigator Hatcher entered the passenger compartment of the Charger and connected a CDR device with the ACM by way of an onboard data port. A personal motor vehicle is plainly among the “effects” with which the Fourth Amendment—as it historically was understood—is concerned, see United States v. Chadwick, 433 U.S. 1, 12 (4) (97 SCt 2476, 53 LE2d 538) (1977), and a physical intrusion into a personal motor vehicle for the purpose of obtaining information for a law enforcement investigation generally is a search for purposes of the Fourth Amendment under the traditional common law trespass standard.

There is no warrant exception that excuses the warrantless search. The automobile exception doesn’t apply because the vehicle was immobilized by the accident and under the control of law enforcement for the duration of the investigation. There were no exigent circumstances for the same reason. The likelihood of evidence being destroyed or going missing was almost nil.

And the reason inevitable discovery does not apply is because there’s no reason to believe investigators would have inevitably discovered the damning evidence legally.

In its order denying the motion to suppress, the trial court concluded that an application for a warrant was inevitable, a conclusion based entirely on its finding that Investigator Thornton “always seeks such a warrant in accidents involving fatalities.” Investigator Thornton, however, was off duty on the day of the collision, and there is no evidence that his involvement in the case was even contemplated until the next day. That a particular officer always seeks a search warrant in similar circumstances cannot logically show that a warrant was inevitable at a point in time prior to that officer becoming involved in the case. Equally important, Investigator Thornton did not actually testify that he always obtains warrants in fatality crash investigations. To the contrary, he said that he always retrieves ACM data in such investigations, most commonly without a warrant at the scene of the crash. Investigator Thornton testified that, if ACM data could not be retrieved at the scene, he would seek a warrant, but that hardly suggests that warrants always or usually are sought, considering that the routine practice is to retrieve data at the scene without a warrant.

Warrants are far from “inevitable” in crash investigations, according to officers’ own testimonies. Applying this exception — as the lower court did — is clearly erroneous. It gives the officers credit for work they almost never do.

With the lower court’s decision reversed, the warrant is suppressed. In Georgia, cops need warrant to search vehicle black boxes. This shouldn’t be that tough to comply with. Crashes severe enough to warrant criminal charges will usually result in immobilized vehicles and a large police presence. Time is on law enforcement’s side. And if it has the time, it can certainly get a warrant.

Filed Under: , , , ,

Rate this comment as insightful
Rate this comment as funny
You have rated this comment as insightful
You have rated this comment as funny
Flag this comment as abusive/trolling/spam
You have flagged this comment
The first word has already been claimed
The last word has already been claimed
Insightful Lightbulb icon Funny Laughing icon Abusive/trolling/spam Flag icon Insightful badge Lightbulb icon Funny badge Laughing icon Comments icon

Comments on “Georgia Supreme Court Says Cops Need Warrants To Search Vehicle Crash Data Recorders”

Subscribe: RSS Leave a comment
This comment has been deemed insightful by the community.
Richard M (profile) says:

Because getting warrants is so hard...

Judges rubber stamp warrants all the time with little or no push back to the cops so the only reason we have all these examples of them not bothering is nothing more than them just being plain lazy and not caring about the laws or the Constitution.

The whole "not all cops" thing gets harder every day.

Ten years ago when I read about the cops doing something awful I gave them the benefit of the doubt. Today I read about them doing something awful and I assume that that they did something awful.

I have no idea of where this is all going to end but I doubt it is going to be anywhere good.

Anonymous Coward says:

Re: Flight and Ship Data Recorders

This speaks to something I was going to ask myself: does this warrant requirement apply to civil investigations of motor vehicle mishaps (such as those carried out by private parties, or the NTSB, or even in some cases by police), or only to uses of EDR data in the criminal context (by a police agency, for punitive purposes)? (Keep in mind that most accidents are just that: accidents, with no connection to any sort of criminal conduct.)

To answer your question, though: this case aside, there is no generic warrant requirement I know of for recorded information in the context of a mishap investigation (whether it be from dedicated crashworthy recorders or device event logging), as these investigations are generally conducted in a civil/non-punitive context. (That’s certainly true when the NTSB is the lead agency as it always is in the aviation world, and may be true in many USCG-led maritime investigations as well, although I’m not as familiar with how the Coasties roll on that front.) This may change if a criminal agency is taking the lead (such as the FBI investigating a crime against air security), but that’s pretty rare to see.

Anonymous Coward says:

Re: Re: Flight and Ship Data Recorders

When you throw in maritime and aviation black boxes, you’re stepping into the commerical regulations which are different. Commercial entities ( maritime, aviation, rail, and even trucking) have different regulations about data access and searches. For example, as a truck driver I’m subject to DOT inspections that may result in civil ir criminal penalties for violations, no warrant required

Anonymous Coward says:

Re: Re: Re: Flight and Ship Data Recorders

Mostly, yes — commercial rules are indeed different, and for very good reason, as you point out. Some privately owned aircraft (mostly larger ones subject to 14 CFR 125 large aircraft rules, and some turbine bizjets in private hands) do have CVRs and/or FDRs, though, and the NTSB’s Recorders Lab has gotten pretty good at pulling data off of navigation/display/monitoring/… systems found in light aircraft involved in mishaps even if the systems were not intended to provide a crashworthy event recording function.

Hugo S Cunningham (profile) says:

Warrants *ought* to be rubber-stamped in a case like this...

as long as the investigation is limited to the last few minutes to determine the cause of the crash. If the cops then go through days of back history in order to trace the car’s previous movements, of course, then that is a different question demanding a more adversarial warrant hearing.

Anonymous Coward says:

Re: Warrants *ought* to be rubber-stamped in a case like this...

The "in a case like this" addition means it’s not a rubber stamp. Judge looks at the warrant application, sees that it is a "case like this" and issues the warrant. Rubber stamping, which is what most judges do, involves see warrant application, issue warrant without confirming that the application involves a "case like this"

Anonymous Coward says:

LEO's are just waiting for the bluetooth interface...

Just wait until your ‘vehicle’ (be it car, boat, or plane) is designed to ‘give up’ the details to anyone with the appropriate bluetooth data retrieval tool (handed out for free to all police departments most likely…)

Sure it sucks for privacy, but, "if you have nothing to hide, then you have nothing to fear, right cops?"

Andrew D. Todd (user link) says:

Post-Event Investigation vs. Pre-Event Safety

When a computerized automobile starts recording information to that extent, the manufacturer, and its agent, the onboard computer, becomes responsible, by way of product liability, for making good use of this information. Determining what happened and preventing it from happening are two sides of the same coin.

I understand that in England, they are calling for cars to be fitted with governors which use GPS to identify the place, then look up the speed limit for that place, and apply it. Under this, system, you can only drive at motorway speeds on a motorway. No speeding tickets, it’s just that the car won’t go any faster. The road in this particular case, Mobley v. The State, is described as one with driveways, which would imply a speed limit of 45 mph or less, and the car was going 100 mph. I think you would find that there were no reasonable grounds for allowing the car to go remotely that fast on that kind of road, and when GPS gets incorporated into the "Minimum Equipment List" the manufacturer would be liable for having failed to fit the car with a context-sensitive governor.

The engineering decision to set a given speed is not only premeditated, but potentially rises to the level of conspiracy. In other words, the speed limit, and any thing else within the computer’s reach, has to be obeyed exactly, to the tenth of a mile per hour— or else. There is no “give” in the situation. When Volkswagen tried to "fudge" the pollution requirements, it got most ignominiously caught. The kind of offense a driver could realistically commit in these circumstances would be to obtain and install a bootleg computer, which allowed what the manufacturer’s computer did not allow.

Of course, the presence of the original computer would be routinely checked in safety and air-pollution inspections, and it might be tied into cryptographic systems, so that keeping the car registered, with a bootleg computer installed,might become a major effort. This act of unlawful modification would be the point at issue, and enforcement would focus primarily on businesses dealing in such parts.

The lapses which a driver might fall into, which are subtle enough that the machine cannot detect and immediately correct them, will also be those which cannot be proven beyond a reasonable doubt. For example, did the driver recognize an object as a pedestrian, and not a piece of wind-blown trash? Even if you have a camera recording showing what the driver saw, it is practically impossuble to reproduce his mental state during the, say, five seconds leading up to the accident.

Anonymous Coward says:

Re: Post-Event Investigation vs. Pre-Event Safety

When a manufacturer makes what appears to them to be "good use" of this information, they become liable, by way of product liability for how they use this information.

In my city, numerous tunnels and elevated motorways coincide with surface level streets. If I’m in a tunnel with a motorway speed limit, I don’t want to be arguing with my car over whether or not I should be travelling at motorway speeds or at local suburban street speed. That’s a serious risk to my safety and to that of everyone else on the road with me. Whether GPS puts me on the surface or in the tunnel can change from moment to moment, meaning I could be sailing along at motorway speed one moment, with a heavy vehicle riding my tailgate and at the next moment, my car could decide to suddenly apply the brakes, putting me underneath 50 tons of truck, trailer and cargo.

This is even assuming that my GPS coordinates are correct in the first place. GPS can drift a long way from the correct location, producing unreliable results.

Somehow, I doubt that motor vehicle manufacturers want to be liable for either scenario and I don’t want to argue with my car over what kind of manoeuvre might be necessary for my safety and that of the people around me. Until cars are autonomous, the operator should be in control of operating the vehicle.

Andrew D. Todd (user link) says:

Re: Re: Post-Event Investigation vs. Pre-Event Safety

To: Mr Toad of Toad Hall, posting as Anonymous Coward:
(If you lack the elementary good manners to at least adopt a pseudonym, you give other people the privilege of christening you).

Well, of course, GPS stands for Global Positioning System. In its first application, aviation, was intended, not to be perfectly precise, but to automatically cue more local and exact navigation systems, such as Instrument Landing System (ILS), and radar altimeters. GPS is essentially a fallback mode, setting a certain minimum of navigational information.

As applied to automobiles, GPS would naturally be superseded by devices built into the road, such as traffic lights which transmit a radio signal containing their full pertinent information (location, color, number of seconds to go yellow, and then red), speed limit signs which display a limit continually recalculated according to congestion and likewise transmit it, or anti-tailgating radar.

Parenthetically, I should like to borrow an idea from fluid physics, that of Reynolds’ Number, and Laminar versus Turbulent Flow. In these terms, a traffic jam is not an accidental phenomena, but a normal and expected result of the speed limit being too high. A properly managed urban freeway might only support a speed limit of 30 mph, which remaining un-congested. Too many cars, too few lanes. Of course, un-congested speed for the downtown surface streets is apt to only be about 10 mph. The streets were historically built for horses; once buildings are built out to the sidewalk, it becomes economically impossible to expand the road; and attempts to go faster than a horse are ultimately futile.

Being rear-ended is probably the safest mode of being involved in a crash, even one with a much larger vehicle. The seat is designed to support your head and neck from the rear. If you speed up to get away from a tailgater, the kinetic energy of the vehicle increases as the square of the speed, and your injuries in a forward-facing crash are likely to be more severe.

afn29129 David (profile) says:

Older automobiles

I love my older automobiles that don’t have all that crap under the hood. No worries about seat-belts. No worries about data recorders. No worries about over-charged Takata airbags that injure or maim the driver or passanger. No worries about excessive complexity (I can repair the thing myself and don’t have to pay someone $135.00 or more per hour.) And the best part; The 1937 pickup truck only has one taillight, it came from the factory that way, confuses the hell out of rookie policemen who think its an equipment violation… until they learn better.

Add Your Comment

Your email address will not be published. Required fields are marked *

Have a Techdirt Account? Sign in now. Want one? Register here

Comment Options:

Make this the or (get credits or sign in to see balance) what's this?

What's this?

Techdirt community members with Techdirt Credits can spotlight a comment as either the "First Word" or "Last Word" on a particular comment thread. Credits can be purchased at the Techdirt Insider Shop »

Follow Techdirt

Techdirt Daily Newsletter

Techdirt Deals
Techdirt Insider Discord
The latest chatter on the Techdirt Insider Discord channel...