Crumb
c/technology · by KayLeadfoot@fedia.io · 1d

Study: Police Reports Miss 96% of Active Phone Use Before Crashes

https://fuelarc.com/tech/study-police-reports-miss-96-of-active-phone-use-before-crashes/

A new Insurance Institute for Highway Safety study compared smartphone telematics with police reports for almost 17,000 crashes in Connecticut, Illinois, Ohio, and Texas. After excluding passive screen activity such as an untouched navigation display, the data still show a huge gap: phones detected active use before crashes far more often than police documented it.

100 points · 18 comments · view on lemmy.world

18 Comments

WhoIzDisIz@lemmy.today · 34 pts · 1d (6 replies)

There Are Some Big Caveats

Phone use before a crash does not automatically make the phone user responsible for the crash.

The researchers didn’t have fault determinations. Someone could be sitting at a red light on a hands-free call and get rammed by a pickup from behind. Activity 29 seconds before impact also tells us a lot less than somebody actively tapping the screen one second before impact.

CMT additionally uses machine learning to estimate whether its user was driving or riding as a passenger. That classification has reportedly exceeded 96 percent accuracy, which is impressive, but leaves some room for error.

So I would not take these numbers and proclaim that 751 crashes were “caused by phones.” The study doesn’t support that. What the study does support, very strongly, is that active phone use immediately before crashes happens a hell of a lot more often than police reports manage to capture.

Gullible@sh.itjust.works · 16 pts · 1d (5 replies)

As someone who has driven both before and after phones became ubiquitous, you are right, there is absolutely some conflation going on here. With that said, people are also absolutely driving more poorly as a result of distractions. The way that people brake and accelerate is exceedingly telling. I wish phones were 100% barred from cars.

WhoIzDisIz@lemmy.today · 12 pts · 1d (4 replies)

Don't forget my favorite telltale: the gradual drift into other lanes followed by a sudden jerk back into their own (or suddenly switching quickly to the one they were drifting into as a kind of "I meant to do that" cover move).

CADmonkey@lemmy.world · 7 pts · 23h

If they look like drunk drivers, but it's daytime, usually you will find them texting.

Gullible@sh.itjust.works · 4 pts · 1d (1 reply)

My favorite aspect of 2026 is how little time people offer before honking at a green light. Everyone knows that people are fucking around on their phones so it's rare to see a light left green for more than 2 seconds without a little *bep*

WhoIzDisIz@lemmy.today · 4 pts · 1d

True, but to be fair a good number of lights seem to be timed ridiculously - presumably to allow cops to hand out more tickets.

proudblond@lemmy.world · 2 pts · 17h

I’ve definitely witnessed a few people do that swerve, and then do it again a few seconds later because they didn’t put their fucking phone down. Comparing them to drunk drivers is absolutely correct.

NaibofTabr@infosec.pub · 16 pts · 23h (7 replies)

What a creepy Orwellian system we've built, to have this kind of information down to seconds. In the next study they'll be including what direction the driver's eyes were pointed in at the time of the crash.

sydd@lemmy.world · 11 pts · 22h (1 reply)

I drove someone's new car the other day. Glancing to my right for maybe a second triggered a "keep your eyes on the road" notification and alarm. You know, I used to not understand older people's fear of new technology. I get it now.

NaibofTabr@infosec.pub · 5 pts · 17h

Yup. I don't think you can approve of this kind of research without also giving tacit approval to the police state being built around us.

credo@lemmy.world · 2 pts · 18h (2 replies)

It’s voluntary (for now). Not hard for a phone app to see you’re using the device just before the accelerometer detects a crash.

Professorozone@lemmy.world · 3 pts · 15h

All cars built after September 2027 in the US must have surveillance technology on board to detect impaired drivers and restrict the operation of the vehicle if impairment is detected. This includes drowsiness. Which means they can track your eyes. Not a big step at all to not only detect if a phone is in use but to record that data.

NaibofTabr@infosec.pub · 2 pts · 17h

It wasn't voluntary for any of the people who had their rights violated by GM, OnStar and LexisNexis. And we know they aren't the only ones doing this shit, they're just the ones that got caught.

If we lived in a sane society, the people responsible for this shit would be occupying prison cells.

If I, as an individual, took it upon myself to install surveillance devices in peoples' vehicles and collect behavioral data on them, without their knowledge and explicit consent, I would be (rightly) held criminally liable for that. But because this creepy stalking is being done by corporations, apparently that's OK.

9488fcea02a9@sh.itjust.works · -1 pts · 15h (1 reply)

I love how drivers only care about orwellian overreach when it infringes in their right to text and drive and kill people....

NaibofTabr@infosec.pub · 3 pts · 11h

That's quite an assumption for you to make. Why would you believe this to be true about me?

pineapplelover@lemmy.dbzer0.com · 4 pts · 13h

It always makes me curious how they even obtained this data

KairuByte@lemmy.dbzer0.com · 3 pts · 12h

I could’ve sworn there was a study that showed phones are less of a problem than gigantic cars, but that may have been damage/death toll instead of crash rate.

LodeMike@lemmy.today · 1 pts · 22h

From the study

The data — which come from safe-driving apps that drivers voluntarily enroll in — show that cellphone use prior to crashes is undercounted in police reports by a factor of at least 7.

Since police reports do not always record crash times and locations with perfect precision, crashes that occurred within the same 15-minute window and within the same one-mile radius were determined to be a match.

This whole study is worthless. There's also a whole ton of assertions with zero source.