Tesla has formally begun providing a limited number of customers with the eagerly awaited Full Self-Driving (FSD) beta version 12. Several customers have since shared their initial impressions of the software built upon neural networks.
According to Teslarati, the automaker commenced the official deployment of FSD beta v12.2.1 on a subset of customer vehicles as soon as employees received the software version over the weekend. Since then, numerous individuals have begun to share their initial impressions of the new software, and the majority appear to have given the version thus far favorable feedback.
Customer Insights: FSD Beta v12.2.1 Marks Substantial Improvements, but Caution Remains Key
Mark Bixby, a customer utilizing the FSD beta since 2021, describes the upgrade from version 11 to version 12 as "the most significant release-to-release improvement" to date. Bixby mentions several setbacks and things that appeared to remain unchanged in a lengthy Facebook post detailing his experience. However, he also identifies eleven areas of improvement, including turning trajectory, turn signals, low visibility turns, and more.
Additionally, he emphasizes the necessity for drivers to maintain readiness to assume vehicle control despite utilizing the FSD prototype, given that it is merely a Level 2 automated system that necessitates constant driver monitoring.
"While v12.2.1 didn't do anything dangerous for me today, don't be lulled by the amazing improvements," Bixby added. "It's still beta software, you are still the driver, and you still have the liability if anything goes horribly wrong."
Bixby and others have also noted that FSD beta v12.2.1 can execute U-turns effectively; however, Tesla did not have to exert effort to teach the system the maneuver. Instead, FSD beta v12 acquires and executes this capability via training derived from its neural network-based architecture.
Tesla's Latest Software Update Elevates Navigation with Road Closure and Incident 'Marks
Software Update 2024.2.6 has significantly enhanced Tesla's Navigation system by introducing "marks" that indicate road closures and incidents. These "marks" are new features previously exclusive to Apple Maps, Google Maps, and Waze.
Although Tesla's navigation system is among the most responsive and immersive in the industry, it still needs a few capabilities available in third-party applications.
For instance, Waze, Apple Maps, and Google Maps provide more frequent alerts regarding impending road conditions. Constraints such as road debris, police, accidents, and disabled vehicles may be crowdsourced and displayed on the screen by other motorists.
Tesla is now exerting comparable efforts.
Green, a Tesla programmer, claims the navigation system will receive numerous enhancements, such as alerts for road closures and accidents. Additionally, it will enable motorists to select routes devoid of tolls.
In addition to other enhancements, the update incorporates route selection options. Desire a scenic, lengthier journey that does not involve the interstate? That is indeed an option.
Time-sensitive navigation will recommend the route requiring the least driving time.
Tesla's most recent navigation enhancement, speed camera awareness, was implemented in early December 2023. In conjunction with several other enhancements, this update introduced driver monitoring.
Photo: Bram Van Oost/Unsplash


Qualcomm Gains on $60B Amazon AI Chip Deal
MediaTek Unveils 2nm Dimensity 9600 Pro AI Chip
Meta Partners With Shopify to Bring Shop Pay Checkout to Muse AI
Europe can’t achieve space sovereignty alone. Here’s why
OpenAI Urges US-Led Global Standards for Frontier AI
Databricks to Invest $350 Million in Singapore AI Expansion
Physicists zoom into the birth of cosmic rainstorms with new CERN study
SoftBank Launches $11 Billion Bond Sale to Fund OpenAI Investment
Burnham Eyes UK Leadership Role in AI Ahead of Trump Meeting
SoftBank Shares Jump 8% as SB Energy IPO Optimism Builds
Big AI wants to slow down AI research. Is it a safety pause or a strategic retreat?
US Northeast Airports Resume Operations After Telecom Outage Disrupts Thousands of Flights
Guterres Urges Stronger Global AI Safeguards Ahead of UN Assembly
SK Hynix, Intel Discuss U.S. Memory Chip Production Deal
Rosenblatt Starts Nokia at Buy on AI Networking Growth
Westinghouse Targets $50 Billion Valuation in U.S. IPO
Cathie Wood Backs AI Slowdown as Safety Concerns Grow 



