Editor’s note: This is Part 3 of a three-part series from TechUnited:NJ on autonomous vehicles in New Jersey. This piece looks at mobility; Part 1 covered the economic opportunity and Part 2 covered safety.
What if I told you there was a technology available today that would make it easier for New Jerseyans who live with disabilities like blindness, epilepsy and neurodivergence, and for the nearly 400,000 New Jersey households that don’t own a car, to get around? Trenton is close to launching a pilot for that technology. Unfortunately, the bill also contains restrictions that keep it out of the very neighborhoods where these New Jerseyans live.
That technology is autonomous vehicles. Rideshare services already serve many of these riders, and for many people they help. A rider who is blind, though, still gets refused by drivers who will not take a guide dog, a problem so persistent that the National Federation of the Blind took Uber to court over it. The Federation’s New Jersey chapter told the Senate this spring that these vehicles mean “unprecedented freedom, independence, and reliable point-to-point mobility.”
Much of the proposed legislation, S1677, is sound. It sets up a three-year pilot with a state task force, requires operators to report every crash, and makes them log 50,000 miles of supervised testing on New Jersey roads before any car runs without a safety driver. These are the right instincts — but 50,000 miles is a mileage standard, not a safety standard.
The proposed restrictions are a different story. As written, the bill says a driverless car:
- cannot operate in a school zone
- cannot operate in a construction zone
- cannot operate in any area with “high rates of pedestrian collisions,” as decided later by NJDOT
• and, if it is a taxi, like most are, it can only run on “designated highways”
No other state has done this. The states that already have driverless cars on the road handled the same concerns differently:
- Pennsylvania requires operators to submit a safety plan showing how their vehicles handle school zones, work zones and emergency scenes, and PennDOT can suspend an operator after a serious crash.
- California approves a specific operating area for each company and expands it as the safety record earns it.
- Arizona, Texas, Florida and Georgia set statewide rules with no geographic carve-outs at all.
None of those states drew the map for the companies in advance, which is what New Jersey would be doing. We ran the numbers on what the school-zone ban alone does, using the 1,000-foot distance New Jersey already uses for school zones:
- Newark: 43% of the city’s neighborhoods are off limits
- Jersey City: 36%
- Trenton: 35%
That is before a single construction zone or pedestrian corridor is considered, and before NJDOT has told anyone how large a school zone is. Nobody can build a service on that map.
The provision that troubles me most is banning driverless cars from any area with high rates of pedestrian collisions. In New Jersey, that list includes our largest cities and the state highways that run through them: Newark, Jersey City, Paterson, Elizabeth, Route 1&9, Route 21, Route 130. Those are the same places where 25% to 40% of households own no car. They are where the older resident who gave up driving at night lives, and where the person who is blind or uses a wheelchair is most likely to depend on someone else for every trip. They are also the places where the safety data says this technology helps the most, with more than 90% fewer crashes that injure pedestrians.
Banning the safer driver from the streets with the worst pedestrian records protects no one. The human drivers who are causing those crashes stay right where they are, and the New Jerseyans who cannot drive are told that this pilot was never meant for them.
New Jersey does not travel in tidy, pre-approved zones. A trip from home to work to daycare to the pharmacy crosses town lines, county lines, arterial roads and side streets without a second thought. A service that has to refuse any route touching a school zone or a busy corridor is blocked from the start, and the New Jersey Chamber of Commerce told the Senate committee the same thing: “This is particularly going to block out urban zones, suburban and mixed-use communities, where the demand for these types of services is often the greatest,” its director of government relations, Amirah Hussain, testified.
Traffic is the last piece, and it is the one every New Jerseyan feels. The average commuter here loses 85 hours a year to congestion, more than double the national average of 35. That is three and a half full days a year spent going nowhere, and we pay for it in gas, in stress, in missed dinners and in the hours of our lives we do not get back. Driverless cars will not erase congestion, but they do take the worst human habits out of the flow: tailgating, weaving and panic braking.
Field experiments have shown that a single automated vehicle can smooth the driving of up to 20 cars around it, and a recent highway trial in California cut hard-braking events by 85% with only a handful of automated cars in the mix. Researchers have found that once a small share of the vehicles on a road drive this way, stop-and-go waves begin to dissolve.
The hundreds of millions of miles behind the safety record Aaron Price described in the last piece were logged on city surface streets, in dense, unpredictable places full of pedestrians, cyclists, delivery trucks and double-parked cars. That is where these systems have been trained and proven, and it is where our congestion lives. The bill names no roads at all. It hands the map to a state agency and fences off the busiest parts of it before a single car has been tested here. It is hard to prove a technology can move a state when the rules hamstring it before it gets off the ground.
Criticism without a better idea is just noise, so here is what a pilot built to work would look like:
- Replace the 50,000-mile trigger with a measurable crash-rate benchmark using the public data the bill already requires, and make crash reporting a national standard.
- Replace the geographic bans with a performance standard: a clear, evidence-based bar an operator must meet to run without a safety driver, and continuous public reporting on every mile, every crash and every disengagement.
- Grow by performance. Start operators in a defined area and expand street by street as the safety record earns it, with the task force the bill already creates empowered to pause anything that is not working.
That is how responsible deployments have scaled in places like Phoenix and San Francisco, and it is far stronger accountability than a person in a seat who may be just as distracted as the rest of us.
The Assembly has not yet taken up its version of the bill, A3968. That is the moment to fix this. As written, the bill guarantees the pilot fails, and then lets everyone conclude the technology was not ready, when the truth is the rules never let it try.
Across these three pieces we have argued that autonomous vehicles are an economic opportunity New Jersey should seize, that they are measurably safer than we are, and that three sentences in the bill on the table would squander both.
Strip away the policy language, though, and this comes down to people:
- the 582 families who lost someone on our roads last year
- the older resident who has quietly stopped driving at night
- the person with a disability for whom “just drive there” was never an option
- the commuter who loses three and a half days a year to gridlock
They are the reason to get this right. The future of this technology in New Jersey is not something that has to happen to us. It is something we get to shape, if we write rules worthy of the moment. Let us decide to lead.
MJ Durkin is the chief operating officer of TechUnited:NJ.
