Ray

The driver of the parked vehicle ahead didn’t check traffic from behind and swung his door open all of sudden, which scared me. FSD swerved left to avoid hitting it.

What do those who claim Bibi is dead say about this opening speech?

Not sure why FSD 14.2.2.5 came to a complete stop in this construction zone—possibly reacting to the mobile flashing yellow light? That said, it did recognize and respond appropriately to the construction worker’s hand signal to proceed.

Watch till the end to see how FSD 14.2.2.5 interprets a truck driver’s hand gesture signaling a merge into a single lane—and responds accordingly.

Not only does FSD understand that my intended route is closed, it also recognizes the detour sign—specifically the arrow indicating an immediate right turn—and responds accordingly. Technically, I could’ve made a left turn at the intersection and been rerouted to my destination. But FSD sticks to what it sees and follows the posted guidance in real time.

At NVDIA GTC 2026, Jensen Huang announced the ChatGPT moment of self driving cars has arrived: “As you know, we've been working on self driving cars for a long time. The ChatGPT moment of self driving cars has arrived. We now know we could successfully, autonomously drive cars. And today we are announcing four new partners for Nvidia's robotaxi ready platform, BYD, Hyundai, Nissan, Geely, all together, 18 million cars built each year, joining our partners from before - Mercedes, Toyota, GM. The number of robotaxi-ready cars in the future are going to be incredible. And we're announcing also a big partnership with Uber.”

@GuruLidar 从未见过那高的楼倒下那么完整, 跑出的人都在说英语🤔

@ray4tesla 几百公斤弹头能有这个效果 这帮人完全没有物理常识😂😂😂

IRGC couldn’t crank out more convincing AI videos including this one.

In a recent launch of its VLA 2.0, XPeng CEO He Xiaopeng commented on the current state of L2 ADAS in China, noting that most systems—aside from XPeng’s VLA 2.0—still rely on HD maps and hard-coded rules. Here is his speech: “We know that many L2 driver-assistance manufacturers have written countless rules specifically for China’s major cities—Beijing, Shanghai, Guangzhou, and Shenzhen—because most media test drives take place there. So when their software is brought to other parts of the world, they have to remove those rules and rewrite them from scratch. But the problem is: they don’t understand the local regulations, and they don’t know how drivers behave locally—how they change lanes, make turns, park, or overtake. As a result, they face tremendous difficulty. This is especially true because they rely on localized high-definition maps, particularly in complex scenarios like intersections. All of this leads to a situation where today’s L2 systems may appear to perform well, but their upper limit is actually very low.”

@JerryLiu112 No word yet.

@ray4tesla Will Elon go with him? It looks Trump won’t bring a lot of business leaders with him this time.

Trump’s official visit to China in two weeks could likely “coincide” with the rollout of FSD v14.

@rechtseflikker Yes

@ray4tesla and data centers in space

Elon posted yesterday about the buildout of Tesla Terafab in a week as Tesla projected in Q4 earnings call the enormous future demand for AI chips—far more than current suppliers can provide. Here are the core reasons. 1. Tesla will need massive amounts of AI chips Tesla’s future products depend heavily on custom AI processors: •Full Self-Driving (FSD) / Robotaxi •Humanoid robots (Optimus) •AI training systems and data centers •Next-gen vehicle computers All of these require specialized chips designed by Tesla. Tesla already designs its own silicon, but fabrication is outsourced to companies like TSMC and Samsung. Even if those suppliers run at full capacity, Tesla believes they still won’t be able to supply enough chips in the future. 2. Remove a major supply bottleneck Elon Musk has said chips could become the main constraint on Tesla’s growth in the next few years. A TeraFab would allow Tesla to: •Produce chips internally •Avoid shortages •scale AI products much faster This follows Tesla’s strategy of vertical integration—similar to its Gigafactories for batteries and vehicles. 3. Control the AI compute layer Tesla sees chips as the foundation of its AI ecosystem. Owning a massive chip factory would give Tesla control over: •chip architecture •manufacturing schedule •cost •performance optimization for its software This lets Tesla tailor silicon specifically for its AI stack instead of relying on general-purpose chips. 4. Scale to extreme production levels The planned facility could start around 100,000 wafer starts per month and potentially scale much higher, putting Tesla among the world’s major semiconductor producers. That scale is needed if Tesla eventually deploys: •millions of robotaxis •millions of humanoid robots •massive AI inference infrastructure 5. Strategic independence Another reason is geopolitical and supply-chain risk. Today most advanced chip manufacturing is concentrated in Asia. By building its own fab, Tesla would reduce dependence on external suppliers and geopolitical disruptions. In simple terms: Tesla thinks the future is AI + robotics at massive scale, and that future will require more chips than the global supply chain can reliably provide. So it wants to build a “TeraFab” — essentially a Gigafactory for AI chips.

@Poxenium @grok That’s the path for Tesla to walk on as it is known for vertical integration in manufacturing processes.

@ray4tesla It will be unforgivable if they will not make their own RAM and SSDs. Market is destroyed. Many will not forgive AI companies for this. @grok will they makee RAM at least?

Tesla haters love to call Tesla community a cult. They clearly haven’t seen what a real automotive brand cult looks like. This screenshot shows a group chat where multiple people are repeatedly posting the same pledge to join Huawei AITO Z7 car owners group. Here is the pledge: “I voluntarily apply to join the Huawei Z7 car owners group. I will safeguard Huawei’s principles, abide by Huawei’s rules, fulfill the obligations of being a Huawei car owner, and carry out Huawei’s decisions.” Huawei has one of the most passionate fan bases in China. Some users treat supporting Huawei almost like supporting a team or movement. Posting a “pledge” message is a way to signal: •Loyalty to the brand •Identity as part of the community •Support for domestic technology companies

@RobWakes @myleninhasanto3 I’m blocked by this fake account 🤣

@myleninhasanto3 for a minute I thought you were @ray4tesla

The future of AV technology is here. It’s great to see this administration taking a proactive stance to support U.S. innovators and keep the momentum of innovation moving forward. 🚗

History made — the first Model 3 drives itself autonomously on public roads with no steering wheel.

Exiting multi-floor garage autonomously is easy for FSD.

@TeslaUltraBull 激励计划让他更加自信😎

@ray4tesla Lol，估计李斌一分钱都拿不到

BREAKING: NIO is the first Tesla competitor to take a page from Tesla’s playbook by introducing a performance-based compensation package for its founder and CEO, William Li. Here are the key details of the package: Recently, NIO Group submitted a report formally disclosing that its board of directors has reviewed and approved the 2026 equity incentive plan. Under this plan, NIO will grant approximately 248 million Restricted Stock Units (RSUs) (specifically 248,454,460 shares) to the company’s Chairman and CEO William Li. If all performance targets are met and the total number of publicly traded U.S. shares remains unchanged, the RSUs could be worth about $12 billion at the highest target level—equivalent to about 82.386 billion RMB. This scale is considered a major equity incentive for a core executive in China’s auto industry. Specifically: •When NIO’s U.S. market capitalization exceeds $30B, $50B, $80B, $100B, and $120B, one-tenth of the shares will vest to William Li at each milestone. •Similarly, when NIO’s net profit exceeds $1.5B, $2.5B, $4B, $5B, and $6B, another one-tenth of the shares will vest at each milestone. If both conditions are met simultaneously—that is, NIO’s market cap exceeds $120B and net profit surpasses $6B—all remaining incentive shares will fully vest. At that point, the value of the stock incentive granted to William Li could exceed 90 billion RMB. As of the U.S. market close on March 12, 2026, NIO’s total market capitalization was approximately $13.981 billion.