Driving Big Brother

The government plans to release new rules for controversial car black boxes this summer, according to a spokesman for the National Highway Traffic Safety Administration.

Contrary to expectations, the rules don't require automakers to install the boxes in every car, but they do require the boxes to record a minimum of 29 pieces of data, more than most black boxes currently record.

In a nod to privacy concerns, the rules require automakers to disclose in the owner's manual when a car has a black box and why it's there. But privacy advocates say they're disappointed that the rules don't limit the amount of data the boxes can record or address concerns about how recorded data can be collected or used.

In New York this year, a defense attorney challenged the admissibility of information gleaned after police removed data from a defendant's black box before obtaining a search warrant.

A judge ruled, however, that the seizure was legal and that the driver had no reasonable expectation of privacy since he'd been on a public highway and exposed his driving behavior to anyone watching.

"Essentially what (the NHTSA) has done is encourage more data collection without a corresponding increase or concern for privacy protection," said Chris Hoofnagle, director of the Electronic Privacy Information Center's West Coast office.

Car black boxes, also called electronic data recorders, or EDRs, have a computer chip that records data about a car and driver's actions before and during a crash. EDRs generally begin recording when the device detects an abrupt deceleration in the car, signaling that a crash may occur. Currently, the data captured varies among EDRs. Some record only vehicle speed and data about air bag deployment. Others also log whether the driver braked before impact, used turn signals or was buckled in.

EDRs were designed to help automakers build safer vehicles. But manufacturers have used the data to defend against product liability claims. Police investigators have also increasingly been using the data to charge drivers with speeding violations and serious crimes. And insurance companies want the data to dispute unwarranted claims and tie policy rates to driving behavior.

Privacy advocates and consumer groups oppose allowing data collected for safety purposes to be used for other purposes, especially when most drivers are unaware that their cars have boxes that can be used as evidence against them. They also question whether the data is accurate, since few tests have been conducted to establish its reliability.

EDRs use proprietary technology, so car owners can't see the software code to determine if the devices are doing more than manufacturers say they're doing. Automobile owners can't access the recorded data without a $2,500 software kit.

Black boxes have been in cars since 1974, when General Motors installed them to help deploy air bags. In 1994, GM began installing more sophisticated EDRs to record data. Now 15 percent of light cars and trucks have EDRs, including 65 percent to 90 percent of 2004-model light vehicles, according to the NHTSA. But not all automakers disclose the presence of EDRs in their cars.

Researchers and automakers say EDRs are invaluable for improving the safety of cars, roadways, bridges and guardrails.

About 6 million crashes occur annually in the United States, according to the NHTSA. The leading cause of death in children and young adults, they cost the country an estimated $230.6 billion a year.

"From a research point of view, there is absolutely no question in my mind that (the data) will lead to safer air bags and cars," said Clay Gabler, professor of mechanical engineering at Virginia Tech's Center for Injury Biomechanics, which began examining car black box data in 2001. Gabler said the data has already helped engineers rethink how highway guardrails are constructed to protect people who crash into them.

But Eric Skrum, spokesman for the National Motorists Association, questions the safety justifications for EDRs.

"If they really were doing this for safety research, they wouldn't have to have it in every vehicle or in vehicles where the owner isn't even aware it's in the car," Skrum said.

EDR proponents also say the devices help victims' families seek restitution against reckless drivers, especially when there are no witnesses to a crash.

When a 77-year-old woman in Texas drove her Cadillac through a post office's plate-glass window and killed an occupant inside, she claimed the car accelerated on its own. Her car's black box belied her story and helped the victim's family win a wrongful-death suit. Similarly, when two cars collided in Montreal in 2001, killing one driver, the surviving driver blamed the deceased driver for speeding. But data from the survivor's black box revealed that he'd been the one driving 80 mph in a 50-mph zone.

Despite the value of EDRs, few tests have been conducted to establish the accuracy of the data. In at least one case, an EDR seems to have recorded faulty information.

Maine Gov. John Baldacci was involved in an accident in his state-owned SUV when the car, driven by a state police detective, hit an ice patch while passing a slower car and caused both vehicles to spin off the highway.

Although the EDR indicated the SUV was traveling 71 mph before the air bag deployed, the detective driving the car claimed the speedometer showed 55 mph, a number that was closer to the 55-65 mph that police investigators estimated based on physical evidence at the scene.

The EDR also indicated that the governor wasn't buckled in, although Baldacci and his driver disputed this. Medical personnel who treated the governor also said his injuries, which included a broken rib, were consistent with someone wearing a seat belt.

Limited tests of EDRs have indicated that if power to an EDR is lost during a crash the device may not record all data or could falsely record seat belt data. The vehicle's speed can also be recorded inaccurately if the car is airborne during an accident, rolls over or loses a wheel from the drive axle. And in at least one case, researchers discovered a programming bug that caused an EDR to falsely record brake information in a particular car model. EDR download reports now include a warning about the glitch for crash inspectors.

Robert Breitenbach -- director of the transportation safety training center at Virginia Commonwealth University, which conducted a study of EDRs -- said investigators should never rely solely on EDR data.

"You really need to do a thorough investigation of the physical evidence and just use it as another tool," Breitenbach said.

He added that people who argue against using EDR data for privacy reasons forget that a lot of information they record can already be gleaned from a car without the driver's permission.

"I can look at your headlights and taillights if you're involved in a crash," he said. "Those lights will tell me whether your lights were on or off in a crash. If they were off and it was dark, that could lead you to a conviction of reckless driving."

Regardless, last July, California became the first state to address questions about access to EDR data when it passed a law that prohibits anyone from accessing EDR data without a car owner's permission or a subpoena. The law contains a loophole, however. If an insurance company assumes ownership of a car wrecked in an accident, the company gets possession of the EDR data as well.

About a dozen other states have introduced similar legislation regulating how EDR data is used or requiring car makers to disclose the presence of EDRs.

Rae Tyson, spokesman for the NHTSA, said his agency recognized the privacy concerns over EDRs, but said the agency didn't have the authority to rule on them and that Congress or the courts would have to resolve them.