A driver can agree to a service without understanding the business built around the information it collects. That gap should concern anyone who thinks a confirmation button settles the question of consent.
In January, the FTC finalized its GM and OnStar order, settling allegations concerning location and driving-behavior data collected and shared without adequate notice and consent. The order includes a five-year prohibition on disclosure of those categories to consumer reporting agencies and longer-term requirements involving consent, access, deletion, and consumer choices, with specified exceptions.
That is a specific enforcement outcome. It is not proof that every connected-car program works the same way. It is a useful reason to inspect how these programs ask people to agree.
A button cannot do all the work
A meaningful choice should explain the consequence in the vocabulary of the person making it. “Improve your experience” is not an adequate description of sending information outside the company. “Personalization” leaves too much unanswered when a person’s movements are involved.
The explanation should survive an ordinary day at a dealership: interruptions, several documents, an unfamiliar screen, and a customer who needs to get back to work. A disclosure that only functions for a patient specialist is poor product design, even before any legal argument begins.
Revocation deserves the same care as enrollment. People should be able to find their current settings, change them, and understand which functions will stop. Requiring a support conversation for every reversal makes the original choice less credible.
Understand the remedy’s boundaries
The FTC’s final order combines a time-limited prohibition concerning consumer reporting agencies with requirements that run for the order’s longer life. It also includes exceptions, such as certain emergency uses. Describing the outcome simply as a ban on collecting driving data would erase distinctions the actual remedy makes.
Those distinctions have practical importance. An owner may want emergency assistance while rejecting unrelated uses of trip information. A product that presents those preferences as an indivisible package makes it difficult to express the choice the owner actually intends.
The lesson is not that every collection of vehicle information is illegitimate. It is that the purpose, recipient, and consequence need to be explained at the point of agreement. “Connected services” is a product category, not a sufficient description of a specific information transaction.
An enrollment process is a piece of machinery
An interface has inputs, states, and outputs. The input here is a person’s decision. The output is a set of permissions the company believes it has received. When the person’s understanding and the recorded permissions diverge, the process has failed at an important job.
That failure can be examined. Present the enrollment flow to people who have not helped design it. Ask them to explain what they accepted, which uses are optional, and how to reverse their choice. Record where they misunderstand the relationship. Then change the interface and repeat the exercise.
This would not resolve every legal question about consent, and it should not be advertised as doing so. It would produce evidence about whether the product communicates. Companies already test interfaces to improve completion rates. They should be willing to test whether completed enrollment reflects an informed decision.
Defaults deserve as much scrutiny as words
The wording on a screen is only one part of the experience. The order of choices, the state selected in advance, and the effort required to decline can influence what a person does. A prominent affirmative button beside a barely visible alternative communicates a preference even when both options technically exist.
A useful review should record those mechanics. It should also distinguish between a setting the customer deliberately changed and a setting inherited from setup. A saved state is not, on its own, a complete history of how the person arrived there.
Manufacturers have a legitimate interest in completing setup without exhausting the buyer. The solution is to prioritize the decisions that materially affect the customer and explain them in layers. A short plain-language choice can lead to a fuller explanation without making the longer document carry the entire burden.
Make reversal an ordinary function
The quality of a choice becomes clearer when someone changes their mind. A usable system should show the current state, the scope of the change, and when it becomes effective. It should distinguish ending future collection from deleting information already retained, because a customer may reasonably want both while the product offers separate controls.
The same reasoning applies when the vehicle changes hands. The new owner needs a fresh account relationship, and the previous owner needs confidence that remote access has ended. An enrollment designed only for the first sale overlooks a normal part of the vehicle’s life.
These are service requirements an owner can understand without studying privacy law. They also give journalists something concrete to evaluate: the complete lifecycle of permission, from first agreement to final withdrawal.
A testable standard
For future vehicle coverage, a useful exercise would be to ask a buyer to explain the data arrangement after completing enrollment. Where the explanation fails, examine the interface before blaming the buyer.
Consent is supposed to communicate a decision. A company should be able to demonstrate that its product helped the customer make one.