ASS(Actually Smart Summon) is a feature everyone should have in the future. It’s fun, user-friendly, and powerful. In this scenario, I’m getting ready to leave the office, and my Model Y comes right up to me.💯
#actuallysmartsummon#tesla
v14 Lite Release Notes:
– Distilled the intelligence from HW4 V14 into HW3. This allows HW3 to directly learn how to handle scenarios using HW4 V14 as a guide. This process unlocks the improvements that have been made to HW4 including Reinforcement Learning (RL) and offline models for HW3.
– Improved both proactive and reactive responsiveness across a wide variety of categories including navigation handling, merges and forks, pedestrian interactions, traffic lights, and vehicle cut-in scenarios.
– Improved general comfort in nominal scenarios through fewer false slowdowns, smoother steering and more consistent lane centering.
– Introduced parking, unparking, and reversing capabilities.
– Added Arrival Options for you to select where FSD should park: in a Parking Lot, on the Street, in a Driveway, or at the Curbside.
– Speed Profiles are now available at all times, to further customize driving style preference.
Many people think any given ML project is 99% training.
In reality, it’s 50% evaluation, 40% data cleaning, 8% integration, and 2% training.
The first two set the noise floor for learning. No ML magic matters; the model cannot lower the noise floor, as that’s the optimal bound of Shannon encoding of your data.
Thus, not a single day goes by without me thinking about ontology. Even the old labels have to be constantly reviewed.
A Model Y driver started experiencing a medical emergency with chest pain mid-drive & called his son.
His son then remotely rerouted the car – which had FSD Supervised enabled – to the nearest hospital & let them know the vehicle was en route. ER staff were standing by on arrival.
Doctors later confirmed the quick reroute likely saved his life.
It is humbling to consider that if we harness just 1 millionth of the Sun’s power for AI, that will be much more than a million times the intelligence of all of humanity
Tesla has announced that it has submitted its FSD (Supervised) application documents in Taiwan to the Vehicle Safety Inspection Center, and that it has officially entered the examination process in the country. https://t.co/LDuPte0Sbz
I have been using it for triaging our fleet data, improving our training labels, patching security holes, and making suggestive improvements in a wide variety of environments.
It’s more than just a coding agent.
Thanks to the xAI teams for their rapid responses and improvements that have boosted my productivity.
I can’t believe how well Tesla FSD handled this emergency vehicle.
It signalled and pulled over before I saw the police car. I’m not sure if it was doing this all based on cues from the other drivers?
Things you don’t need to worry about when buying a Tesla:
• Oil changes (forever)
• Oil filter, air filter, or fuel filter replacements
• Spark plugs or ignition coils
• Transmission fluid changes (Tesla has a simple single-speed gearbox)
• Exhaust system repairs (mufflers, catalytic converters, oxygen sensors)
• Radiator, water pump, or coolant flushes
• Timing belt/chain replacement
• Fuel pump or fuel injector cleaning/replacement
• Serpentine belt or accessory belt replacements
• Head gasket failures
• Replacing 12V batteries (newer Teslas have lithium batteries)
• Transmission faults or failures
• Emissions testing & failing smog checks
• Overheating from coolant leaks or failed thermostats
• Stopping at gas stations
• Gas price fluctuations (the best)
• Multiple brake pad/rotor replacements (regenerative braking enables brakes to last onwards of 100k-200k miles for some people)
• NO MORE CHECK ENGINE LIGHTS.
• Driving past 3 gas stations because “it’s 10 ¢ cheaper 2 miles away”
• Catalytic converter theft (yes that’s a thing)
• Not dealing with aggressive dealerships and service.
+ many more.
All the maintenance a Tesla really needs is:
• Tire Rotations (if applicable)
• Cabin Air filter replacements
• Brake fluid changes if needed after MANY years
Unlike a gas car that has 2,000-3,000 moving parts in its powertrain, a Tesla has only 20-50 moving parts.
That’s literally 100 times fewer moving parts than a normal gas vehicle. It’s the main reason Teslas (and EVs in general) are so reliable and need almost zero maintenance.
BREAKING: Tesla has officially received approval from the Arizona Department of Transportation to launch its Robotaxi service on public roads in the state!
I just confirmed directly with the Arizona DOT that @Tesla applied for a Transportation Network Company permit on Nov 13th. Tesla met the requirements to operate as a TNC and is permitted as of Nov 17th.
Tesla can now legally run a paid, autonomous ride-hailing service in Arizona. The Arizona TNC permit is not limited to a specific city 😎
全新 Model Y 蟬聯 6、7、8 月台灣電動車銷量冠軍與豪華進口休旅車榜首,並於 8 月取得台灣總市場銷售排行榜(不含商用車)第二位、進口車銷量冠軍,更是進榜前十名唯一的純電車款。
邀請您立即點擊連結預約專屬試駕,親自感受全球百萬車主首選純電 SUV
The New Model Y continues its reign as Taiwan’s best-selling electric vehicle in June, July, and August—also taking the No. 1 position among luxury imported SUVs.
In August, it rose to No. 2 in overall vehicle sales (excluding commercial vehicles), standing out as the only pure EV in the top ten—and the leading imported car.
Click the link now to schedule your exclusive demo drive to experience the world’s most popular all-electric SUV
https://t.co/a7miBP2Kd7
Elon always explains this without a concrete example, and that’s why most people still don’t really get the point, and fail to understand why disagreement would reduce safety. Let me provide one simple real-world example:
At night, LiDAR usually will have difficulty “seeing” a black tire on the road due to low reflectivity, but cameras with higher dynamic range can (even clearer than a human would). Real-world tests in China have demonstrated that a system with sensor fusion would be less likely to avoid a tire on the road during the night at high speed. This is mostly because the system decided to believe LiDAR that there is no obstacle on the road.
Also, recent safety tests performed by DCar in China showed that FSD outperformed every other sensor fusion system in safety testing, further demonstrating the advantage of a system trained with input that has a higher signal-to-noise ratio.