Tesla Introduces Early Eyes-Off Feature; Allows to Text and Drive

Not a Tesla App
Karan Singh

Tesla has quietly unlocked a capability that they announced was coming — the ability to text and drive. Or at least, sometimes, depending on traffic.

In this recent exchange, Elon confirmed that Tesla’s latest FSD release, FSD v14.2.1, now allows users to use their phones “depending on the context of surrounding traffic.” This is a drastic change in Tesla’s driver monitoring, which has often been strict, issuing warnings for even minor glances at the center display.

FSD v14.2.1 now allows users to send a brief text under certain conditions without the driver-monitoring system warning the driver.

By relaxing the driver-monitoring system to allow eyes-off behavior under certain traffic conditions, Tesla is effectively deploying Level 3 autonomy. But there’s a critical catch: unlike competitors who offer similar features in extremely tight constraints, Tesla is not accepting liability. This creates a dangerous gap. The car grants you the privilege of brief distractions, but you retain 100% of the risk.

The Uncertified L3

The behavior that Elon describes - allowing phone use in specific, system-approved contexts is similar to Mercedes-Benz’s Drive Pilot, which is certified as an L3 system on certain stretches of highway in California.

When Drive Pilot is active (under 40 mph in traffic jams, with no inclement weather, a leading vehicle, and no construction), Mercedes explicitly informs the driver they can watch videos or text - and that liability rests with Mercedes for the duration the system is active.

Tesla’s approach with FSD v14.2.1 appears to mimic this behavior - specifically, scenarios FSD is confident it can handle - but without the certification or acceptance of liability. There is no regulatory approval for L3 operation, and there are currently no labels or warnings indicating that the car is now fully confident and in control, and, more importantly, no transfer of liability.

New Icons

Not a Tesla App

A recent discovery in Tesla’s source code uncovered new attention required icons. The three icons represent varying amounts of attention required by the driver. While these icons aren’t displayed to the user in the latest FSD releases, it looks like Tesla may be preparing to show users how confident FSD is in the current situation and what they require from the driver.

Elon’s caveat that it depends on the context is likely a nod to the patchwork of state laws regarding distracted driving. In states like Florida and Texas, the law allows drivers to use handheld devices if the vehicle is stopped, like at a red light or in gridlock.

Tesla has already relaxed driver monitoring when the vehicle is stopped, regardless of the state you are in. In fact, in some other states, you’re considered operating a vehicle even when it is stopped, making device use completely illegal.

This creates a trap for owners. Your Tesla may not nag you for texting in a Los Angeles traffic jam, implicitly telling you it is safe. But if a police officer sees you, you’ll still receive a ticket. The officer doesn’t care about what version of FSD you are running, only that you are holding a phone, which remains illegal.

True Eyes-Off

If Tesla wants to allow eyes-off behavior, it needs to properly implement a safety net - and the autonomy guidelines that define Level 1 up to Level 5 already have a solution: a Minimum Risk Maneuver.

In certified L3 systems, if the fully autonomous conditions end and the driver is too distracted to take control, the system doesn’t just warn you and disengage. Instead, it pulls over to a safe stop and waits for you to pay attention again and continue as an L2 system. 

Don’t Put Yourself in Danger

With FSD v14.1.2, Tesla is effectively putting drivers at risk. By removing the friction of the nag and the threat of a strikeout, they are empowering users to disengage from the road. But unless Tesla formally applies for L3 certifications and backs it up with an insurance liability shield, drivers should treat this newfound freedom with extreme caution.

Keep in mind that Tesla itself still does not operate vehicles with FSD without a safety monitor paying close attention, even in a smaller area like Austin.

According to Tesla’s own stats, the average human-driven Tesla has a minor or major collision every 740,000 miles. While FSD has come a long way and is better than it ever has been, it is still not as safe as a human driver when unsupervised. Just because your vehicle can handle a 2-hour drive to the airport and back without an intervention isn’t a sign that it’s as capable as a human.

If someone witnessed FSD driving for a whole year without any incidents, they may be tempted to say it’s safer than a human. However, according to NHSTA data, the average driver gets into an accident once every 18 years (every 229k miles).

Our safest bet right now is to use FSD responsibly. It offers capabilities humans don’t have, such as viewing all angles at once, but it still requires an attentive driver who’s ready to take over right now.

If you need to check a notification or send a quick text, that should be fine, but keep your phone use limited; even if the vehicle allows for more, you may be putting yourself and your family in danger.

Your car may let you text, but if you crash or if a cop stops you, Tesla isn’t paying the fine. You are.