
Loading summary
A
For the last 100 years, basically ever since the era of Henry Ford, the automobile has been a symbol of American freedom. You just jump in your car, you hit the road, and you go wherever you want, whenever you want, so long as you can afford the gas, the insurance, and as long as your registration and your car inspections are all up to date. However, very soon you can also add to this list whether or not the onboard AI sensors consider you fit to drive. This is actually already a feature in some new car models, but because of a change in federal policy, it's actually getting rolled out across the board for all new cars starting in the year 2027, which is next year. To get a picture of what this feature looks like right now, here's a short video of a guy complaining about his wife's new car over on X.
B
All right, I got to come on here and vent real quick. My wife got a new car through her work. It's a brand new. It's got no miles on it. Brand new 2025 or 2026, I imagine, Subaru Outback. And it's got the feature where it says. Or it's the eyesight thing. So it's got a camera on the dash. If you veer off for two seconds, like if you look away like this, or if you're looking outside, I got mountains around us, right? So I'm looking at the mountains, it'll flash and beep and say, keep eyes on road. Changing my music on Spotify for two seconds and it's flashing, saying, keep eyes on road. I get the idea of it, but this is the dumbest thing I've ever seen in a car.
A
Now, what he's describing in that video appears to be the direction that car surveillance is headed in. If the car's AI decides that you are not responsive enough while driving the car, it can automatically break. It can reduce certain functions, or it can. It can even take over the steering function and put the car over to the shoulder and cut off the engine. Basically, if the AI in the car decides you're not up to driving, it will just take over, shut down, and prevent you from driving. With the implication being, if you think it through, that the AI in the car is always in the process of watching you because in order to figure out whether you're impaired or you've gotten impaired after you began driving you, it needs to always be watching you. And as I mentioned a moment ago, this feature, which has again already been included as a safety feature in some newer car models, will become the industry standard Very soon. That's because back in the year 2021, so five years ago, Joe Biden signed into law the Infrastructure Investment and Jobs Act. And buried within that 22,700 page bill was section 24220, which directed the National Highway Traffic Safety Administration to establish a new federal vehicle safety standard. This safety standard, it requires all passenger vehicles to be equipped with what the government calls, quote, advanced impaired driving prevention technology. Now the deadline for the agency to finalize these rules was late 2024. However, they missed that deadline citing what they called technological complexities, complexities that we'll get to in a moment. However, regardless of them missing that deadline, the policy is now moving forward quickly industry wide with full industry integration projected by again next year 2027. Describing the current state of the situation, you have a trade publication called Motor1 write the following quote, BMW, Ford, General Motors and Toyota have written to the agency stating they are generally supportive of the initiative. But each raised concerns about consumer acceptance and accuracy. Now, consumer acceptance is obvious. You can imagine Americans don't necessarily want to have themselves be constantly watched by the tech in their car. But it's the accuracy portion that's become sticking point in terms of implementation and that become the reason why becomes evident when you actually look at how this technology is going to work. Basically, the vehicle must monitor you constantly without your active participation. And to achieve this, automakers are developing a suite of internal surveillance tools. You have infrared cameras that are pointed at the driver's face. You have biometric sensors in the steering wheel as well as software that's designed to analyze the behavior of the driver in real time. And so let's go through these together, starting with the onboard cameras, the ones that are being developed. They use infrared light to track your eye movement, your blinking rate, as well as the general position of your head. And if the system detects your eyes closing for too long, or if your gaze starts wandering from the road for too long, or if it flags something that you're doing as a sign of fatigue or distraction, then it will take the necessary steps to fix that. Then the second way that they're going to be doing the surveillance is the biometric sensors within the car. These companies are prototyping advanced touch based systems that would be embedded in either the ignition button and or into the steering wheel. And they'll be able to read your blood alcohol content through your skin. It's basically like a chemical sensor that you would have to interact with in order to drive the car, I guess, unless you wear gloves. But even if you do wear gloves, there will be another sensor which would monitor the air inside the car in order to detect alcohol molecules being exhaled by the driver. And then, in terms of the behavioral tracking, the car's onboard computer can constantly measure your steering input, the force that you're using to hit the brake, as well as how you're keeping in your lane. And if your driving patterns match a statistical profile of someone who's intoxicated, the system then triggers an automatic, quote, unquote intervention. And in terms of what an intervention actually means, the new law explicitly dictates that when impairment in the car is detected, the technology must prevent or limit vehicle operation. Meaning, in summary, if the vehicle's AI system decides that you are unfit to drive, it has the authority by force of law to override you and take control of your car. The one you paid for and the one you put gas in, the one you pay insurance for, the one you Pay registration for, etc. It will take it over. This can take either the form of the car refusing to start, as I just mentioned, or if you're already driving, the system could put the car into what's known as a limp mode, which restricts your speed. It turns on the hazards automatically, and when it's able to, it'll pull the car over to the side of the road. However, if all that sounds too science fictiony to you, even with all the recent technological AI advances, well, actually, the National Highway Traffic Safety Administration agrees with you. In fact, the agency released a statement saying that despite many modern vehicles already having one form or another of these driver monitoring systems, that the technology required by the mandate, which would be ubiquitous and would go into every car, is not yet mature enough. They added that the technology available cannot achieve the necessary accuracy which is in and of itself, danger to people's lives. Specifically, in a report that they gave to Congress in February, here is what the agency wrote. Quote, even a 99.9% detection accuracy level could result in millions to tens of millions of instances each year where the technology would incorrectly prevent or limit drivers from operating their vehicles, or fail to prevent or limit impaired drivers from doing so. They added later in the report that some of these false positives could come as a result of something as simple as a driver rubbing his or her tired eyes as they're driving. And the system might register that as a sign of intoxication. Or as another example, the air sensor in the car might misread someone's perfume or cologne as them having an elevated blood alcohol content and then shut down the vehicle. And because of these hurdles, the journalists over at that motor, the motoring website, the industry mag, Motor1, they are questioning the 2027 launch date. Quote, it is only a matter of time before technology is capable of detecting impaired driving, but it likely won't be in new cars starting next year. There are still too many hurdles preventing the monitoring systems from working at a mass scale, which gives lawmakers time to reconsider the mandate. But here's the thing. Despite the technological shortfalls, the legal requirement remains in place for these things to be implemented in all new cars starting next year. Now, there has been an effort to repeal the mandate within the House of Representatives, but that has been stalled in committee as of this recording at least. And barring any sudden reversal in the law, the technology is coming in some form or other. And so I guess we can all look forward to that. Also, just generally, it's worth mentioning that aside from worrying about your vehicle shutting down on you randomly, there's also a much broader issue at hand here, which is the transformation of the car into a government forced data collection center. Critics argue that modern cars are fast becoming surveillance pods, with new cars, the ones that are set to be released in 2027 being the most invasive piece of technology that someone could own, potentially even surpassing their smartphone. Because if you think about it, the sheer volume of data that could be collected during a simple commute is voluminous. I mean, the vehicle records would contain precise GPS locations, the exact routes that you've been taking, your average speed, your braking habits, the fours that you take corners in, and so on. Then you add to it the internal cameras, the microphones, all the sensors, and you basically have a device that not only tracks you everywhere you go, but but also actively watches your facial expressions, monitors your physical biometric markers, and potentially captures every single conversation that you're having. And the question naturally arises, where does all this data go and who ultimately owns that data? In an interview with the New York Post, you had the CEO of a company called the Privacy Bee say the following quote, it's definitely not going to stay in the vehicle. It's tracking all kinds of visual signals, and they're usually sending them up to some kind of automotive cloud. A handful of large automakers take this behavioral data and bury it in your terms of service. You give permission for the automaker to sell that data. They're actively selling it to companies that create an automated risk score on an individual and sell that to insurance agencies. And we've seen many cases of this of automakers sharing driver behavior info with third party data brokers who then turn around, package it nicely together and sell it to insurance companies. In fact, there have been reports from all across the US of people who have experienced sudden massive spikes in their insurance premiums or sometimes just had their insurance policies canceled altogether based solely on the data that was harvested from their own cars without their explicit informed understanding. Although I guess they did technically give consent when they signed that form. And so there is the inherent tension because on the one hand this is a clear expansion of the surveillance state, with the government and the private sector mandating ever more that you need to be watched on a moment to moment basis in your own car, which is just one step removed from your own house. On the flip side, however, I mean you can understand the I guess the benevolent intention here if you believe there is one that drunk drivers are pretty much the worst. They kill people through their negligence. And it would really be amazing if a car just refuse to drive someone who is so inebriated that they can't even walk straight. But then of course there's the issue of false positives. So even with a system that's 99.9% accurate, there's still going to be millions of people, tens of millions of people, or let's say tens of millions of trips in which somebody is inaccurately judged to be drunk. And so let me know your thoughts. Do you agree with this development? Do you think that the potential increase in safety does it warrant the additional invasion of privacy? And also do you agree with my personal thesis that starting very soon, cars that are 2026 and earlier are going to skyrocket in value as people just try to avoid these new AI infused vehicles. Let me know your thoughts in the comments below. I'd love to know them. I'll be reading through the comments tonight as well as into the weekend as the video gets more views. And then lastly, two things in the description box I will throw the the links to all my research notes. You can read through the relevant parts of the law, what the agency said about the technological shortfalls in that report in February, as well as some some industry reports about where the technology's at and how it's going to be rolled out, et cetera. And then please do smash those like and subscribe button so this video can reach ever more people via the YouTube algorithm. And then once the those people come in, they watch the video, they'll leave comments. It'll reach ever more people. It will be a nice and positive cycle and this video can get a million views, hopefully. So again, smash those like and subscribe buttons. Thank you very much.
B
Subscribe.
A
And then until next time, I'm your host, Roman from the Epoch Times. Stay informed and most importantly, stay free.
Host: Roman (Epoch Times)
Episode Date: June 19, 2026
This episode explores the upcoming federal mandate requiring all new cars (starting in 2027) to include advanced AI-driven surveillance systems that monitor driver behavior and can override or disable vehicle operation if impairment is detected. Roman discusses public concerns, technological challenges, privacy implications, and potential industry and legal pushback—framing the change as both an advancement in driver safety and a significant expansion in surveillance.
The requirement originates from the Infrastructure Investment and Jobs Act (2021), specifically section 24220:
Industry generally supports the initiative but is concerned about consumer acceptance and accuracy:
Continuous Surveillance:
System Response:
The National Highway Traffic Safety Administration (NHTSA) reports the tech isn't mature or accurate enough for mass deployment:
Motor1 (industry mag):
The broader issue: cars as “government forced data collection centers”—collecting far more data than smartphones:
Privacy Bee CEO (cited):
The Dilemma:
Prediction on Used Car Market:
Roman sums up the debate: While the goal of reducing drunk driving is praiseworthy, the technology’s shortcomings and broad potential for abuse raise significant concerns about privacy, autonomy, and unintended consequences. The episode ends by soliciting listener opinions—inviting them to weigh in on whether the safety benefits justify such pervasive surveillance, and predicting a surge in demand for “pre-AI” cars.
Links to research and further reading are available in the video description section, as mentioned by the host.