As Tesla’s highly anticipated Robotaxi launch in Austin, Texas, continues to fade into the horizon like a mirage on a scorching summer day, concerns mount that this once-promising prospect has devolved into a haze of uncertainty, threatening to leave investors and customers alike feeling duped.
Despite CEO Elon Musk’s assertion that Tesla prioritizes security, concerns arise that the company may be sacrificing substance for appearances.
It’s all about optics
Elon Musk has been consistently skeptical about the feasibility of fully autonomous driving systems for several years. His monitor file bears the scars of chronic tardiness and shattered promises.
He stated:
“Our goal is to achieve a showcase demonstration of fully autonomous driving capabilities, connecting Los Angeles to New York, with a seamless journey from our LA base to a drop-off point in Times Square.” In New York, the technology allowing cars to park themselves was in place by the end of the subsequent year. Without the need for a solitary contact, alongside the charger.
In retrospect, he predicted that it would take place by the end of 2017. By 2025, however, Tesla still struggled to achieve this goal.
Musk has stated that Tesla would achieve fully autonomous driving capabilities without human supervision by the end of the last decade. It has developed into a habit of constantly gagging on his missed deadlines.
Predictions surrounding Tesla’s self-driving capabilities frequently focus on achieving this milestone by year’s end or, at the latest, within the following 12 months.
Musk has established a specific deadline of June for the rollout of Tesla’s autonomous ride-hailing service, robotaxi.
As Waymo surges ahead, expanding its presence to four cities and logging over 200,000 paid rides weekly, Elon Musk is determined to maintain his momentum by claiming a major victory in autonomous driving, having long touted Tesla as the industry leader in this space.
Not long ago, Elon Musk boasted about Tesla’s autonomous driving capabilities.
Nobody is even shut. There isn’t actually a thorough second glance. Were concerned that the concept is akin to an iPhone – either you fully understand its value or it simply doesn’t resonate, leaving a significant gap.
As Waymo’s expansion gains momentum, it’s becoming increasingly challenging to articulate their trajectory. Despite this, Musk claims the robotaxi launch in Austin will maintain the momentum, despite Waymo having operated in Austin and other cities alongside Tesla’s plans for years.
What drives individuals to put up with their current situation?
The controversy surrounding Tesla’s autonomous ride-hailing pilot in Austin has been characterized as a move that shifts the goalposts on the company’s self-driving ambitions?
Since 2016, Tesla has consistently pledged to deliver autonomous driving capabilities without human supervision on all vehicles manufactured during that timeframe. Elon Musk emphasized that customers who acquired Tesla’s Full Self-Driving package would be able to relinquish control for extended periods, effectively allowing them to “doze off” at the wheel without worrying about safety concerns or needing to intervene when arriving in a new city.
Musk’s recent claim that Tesla has effectively solved self-driving technology through the launch of its “Robotaxi” initiative raises eyebrows, given the stark departure from earlier promises.
Tesla intends to operate its own dedicated fleet of vehicles within a geo-fenced area of Austin, utilizing software tailored for this specific space and supported by “substantial” teleoperation capabilities. This marks a stark contrast with the company’s long-promised plans to deploy unsupervised self-driving technology in customer-owned vehicles, a goal that has yet to be realized since 2016.
According to Musk’s own statements, he has acknowledged that the availability of geofenced areas does not necessarily imply the presence of actual self-driving capabilities.
Musk has reportedly stated that Tesla’s Full Self-Driving (FSD) beta program will initially be launched in a controlled environment in Austin, Texas, with specific restrictions to ensure the service can operate safely within designated areas and avoid uncertain intersections.
We’re planning to restrict access to a specific geographic location using location-based technology. We won’t proceed with taking intersections until we’re thoroughly convinced of its success in navigating the specific intersection. Alternatively, it could opt for a detour around that junction.
While Tesla leverages geofencing to govern vehicles, the company can also employ teleoperation, allowing human operators to remotely control cars and maintain order.
We revealed last year that Tesla started building a teleoperation team.
Regardless of Tesla’s initial plan to debut its robotaxi service on June 12, which has been pushed back to “tentatively” on June 22, the company recently posted a job listing for engineers to work on building a low-latency teleoperation system to operate its self-driving vehicles and robots.
Tesla’s reliance on geofencing and teleoperation constrains its autonomous capabilities to a degree mirroring those of Waymo, contradicting Elon Musk’s earlier claims that it is “actual self-driving” and can be scaled to the entire fleet as initially promised.
‘Paranoid’ about Security
Despite Musk’s claim that Tesla is being “tremendously paranoid” about security, one must interpret his statement with caution.
Despite having previously highlighted this, it’s still essential to reiterate: Waymo tested its self-driving cars in Austin for six months with safety drivers and then an additional six months without safety drivers before launching its autonomous ride-hailing service in the city?
As part of an extensive evaluation process, Tesla conducted thorough assessments of its vehicles with trained security drivers in Austin over the course of several months. Musk unexpectedly unveiled in late May, mere weeks prior to the planned rollout, that the company had started piloting its autonomous vehicles without safety operators on board.
Despite widespread anticipation, autonomous Tesla Robotaxis were spotted for the first time last week.
Since then, only two confirmed Tesla vehicles without human drivers have been spotted undergoing testing.
Several of these vehicles have been spotted with Tesla employees in the front passenger seat. While Elon Musk maintains that there is no safety driver on board, the “test passengers” are actually equipped with an override system, allowing them to regain control of the vehicle in the event of an emergency.
While they essentially operate as “security drivers”, they’re seated in the passenger position rather than behind the wheel.
According to state regulators’ listing, Tesla remains in the “testing” phase, with no mention of safety drivers.
Tesla’s head of self-driving, Ashok Elluswamy, recently shared a photo of the company’s “robotaxis” undergoing examination in Austin – a conveniently cropped image that sparks curiosity about the project’s progress.
The photograph captures the absent passenger seat of the lead vehicle, possibly featuring a Tesla employee, alongside the driver’s seat of the following car, likely occupied by a driver, as witnessed in Austin during the past week.
Currently, there is no way to determine precisely when and how security passengers and remote operators intervene in self-driving vehicles’ operations.
Despite touting “miles between disengagements” as a key indicator of progress, Tesla has never publicly disclosed any information about its intervention or disengagement rates in relation to its autonomous driving and advanced driver-assistance systems (ADAS) applications?
With previously reported findings suggesting that some of the most valuable insights stem from collaborative endeavors, we see that crowdsourced efforts can indeed yield remarkable results. Musk had previously disseminated and misinterpreted the dataset in a somewhat misleading manner.
Currently, the information for the most recent updates (v13.2.8-9) on Tesla’s newest hardware (HW4), reportedly used in their “robotaxis” in Austin, shows a significant milestone of 444 miles between critical disengements.
Will automating driving decisions without human oversight pose an unacceptable risk of accidents every 444 miles?
Tesla is currently engaged in a legal battle against companies seeking access to its self-driving car crash data.
Efforts are underway to revisit and further discuss the planned deployment of Tesla’s autonomous taxi service, or “robotaxi,” in Austin.
The Daybreak Venture highlighted the potential risks posed by Tesla’s forthcoming robotaxi fleet, illustrating the issue through a demonstration of how Tesla vehicles fail to brake for school buses with activated stop signals using the latest public Supervised Full Self-Driving (FSD) software version 13.2.9, potentially putting children in harm’s way.
Musk has consistently emphasized that the vehicles employed for its robotaxi service in Austin are the same ones currently delivered to customers, as exemplified by this test ride.
Despite these claims, a newly developed, tailored software application, allegedly optimized for Austin’s unique needs, is said to facilitate increased productivity through additional parameters. Despite the lack of a clear strategy, there is no confirmation available, since Tesla has yet to release any relevant information.
Electrek’s Take
I can’t lie. I’m becoming increasingly engrossed in this matter. It’s difficult to accept that Musk and Tesla could guarantee the safety of such an ambitious project in its current form.
As I previously noted, I think Tesla’s Full Self-Driving (FSD) technology would be an extraordinary product if marketed as a high-end Advanced Driver-Assistance System (ADAS), rather than promising something called “Full Self-Driving” with the implication that it will eventually become unsupervised.
If Tesla fails to honour its commitments to customers regarding HW3 updates, it would assume a significant legal liability, given prior confirmation of these guarantees to homeowners. Additionally, enhanced security features could be implemented, thereby reducing driver reliance on autonomous technology and fostering a greater sense of vigilance.
As the specter of failed promises looms large, it’s no wonder that Tesla is pushing hard for a swift launch out of its Austin facility.
As John Krafcik, Waymo’s former longtime CEO, once remarked about Tesla’s endeavors
Elon Musk’s eagerness to succeed in the realm of autonomous driving has prompted him to seize an opportunity for victory, capitalizing on the fervor of his devoted Tesla shareholder base by tantalizing them with fleeting glimpses of the “Tesla robotaxis” that have long been promised.
Acknowledging his earlier statements, he recognizes that Tesla’s approach may not be more scalable than Waymo’s despite lower hardware costs per vehicle. The hardware cost is virtually inconsequential when considering the larger expenses of teleoperation, business expansion, insurance premiums, and other financial burdens.
Despite the elaborate strategies surrounding this endeavour, it’s increasingly apparent that Tesla lacks the necessary preparation.
Will Musk permit the June deadline to expire, adding another significant “L” to his company’s autonomous driving credentials, or will he choose to push ahead with a potentially hazardous feature, acknowledging its numerous limitations?
Given the turmoil within federal agencies and potential conflicts of interest among state authorities tied to Elon Musk’s interests at Tesla, it’s unclear whether regulatory bodies can be relied upon for fair guidance in this situation? Despite their best intentions, they probably should.
