AITO M9 Hits Pole and Catches Fire, Driver Killed! Huawei Customer Service Learns from the Cyberspace Administration to Tell You Not to Believe Rumors? The Pole Didn't Bend, So Impact Speed Was at Most 50 km/h; 139.1 km/h Is Real but Cherry-Picked; Seres Uses Car Purchase Payments to Earn Bill Discount Spreads; Sues Netizen for Millions in Damages; 'Far Ahead' Has Become Selling Matches!

An accident occurred in Shenmu, Yulin, Shaanxi, where an AITO M9 crashed into a lamp post and caught fire, killing the driver. AITO User Service issued an announcement stating that the vehicle speed reached 139.1 km/h and was under manual driving, sparking strong public doubts regarding the completeness of data disclosure, the PR attitude of car companies, and the supply chain financing model.

Video link:

Full Text:

See things as they really are, refuse fragmented perception. This piece of shit Huawei has really been making a lot of noise recently. Just last Saturday, Huawei fucked over another one of its own fans to death. In the future, stop calling it "far ahead" or whatever; just call it "Huawei the match seller." The incident happened in Shenmu, Yulin, Shaanxi, where an AITO M9 either lost control or something else happened. The video shows that after rushing through that intersection, it slammed directly into the square traffic pillar, and then caught fire in less than two seconds. That fire caught faster than the stove my great-grandmother lit, and the driver inside died on the spot. Truth be told, this match-selling little boy Huawei isn't completely without merits—this fucking piece of shit car sells for a fucking 480,000, so it's definitely not aimed at screwing over the working class. Moreover, no one who isn't a brain-dead Huawei fanboy would buy this fucking thing. Being rich and a Huawei fan are the necessary and sufficient conditions to be screwed over by Huawei. But truth be told, if you're selling matches, just sell matches; robbing money is one thing, but you fucking take lives too!

Just four days later, the AITO User Service account posted a 300-word statement, saying that at the time of the accident, the car crashed at a speed of 139.1 kilometers per hour, but the speed limit for that road segment was 50, and it was under manual driving status. What were they saying? Ah, that there's nothing wrong with the car. Then, they categorized battery explosion/deflagration and smart-driving collision as false information. What was the last sentence? It said "do not believe or spread rumors, and jointly maintain a clean online environment." Fuck, do they fucking think they are the Cyberspace Administration of China or what? A car seller, four days after someone burned to death, this match-seller Huawei uses the sentence pattern used by the Cyberspace Administration of China to lecture netizens to lecture consumers. Today, let's look at how this fucking idiot match seller was fed bite by bite into an arrogant monster by three things: black-box technology, discourse hegemony, and supply chain finance.

Let's first take a look at the structure of this fucking announcement. Even though this statement is several hundred words long, the central idea actually boils down to four paragraphs. The first paragraph says that after the crash, a fire broke out, the eco system triggered the rescue center, and they called 112. The second paragraph says the vehicle was in manual driving mode, and when passing through the intersection, the vehicle speed was 139.1, with a speed limit of 50. It implies that because the speed was too high, it hit that lamp post. The fire originated in the front of the vehicle, and no abnormalities were found with the vehicle prior to the incident. Fuck, 139.1! Given the quality of Huawei's fucking cars, if it were really driven at that speed, wouldn't it fucking smash out atomic nuclei! The third paragraph says the fire was caused by an electrical short circuit and fluid leakage resulting from the collision, and that they are currently cooperating with the investigation, with detailed conclusions subject to the official final investigation findings. The last paragraph is just that "clean cyberspace" line. Everyone can go count its subjects. The subject of the first paragraph is the vehicle and the system; hey, it mainly just claims there is nothing wrong with the system and the vehicle. The subject of the second paragraph is the vehicle. The subject of the third paragraph is "our side," this match-seller. The subject of the fourth paragraph is "us and everyone." Across the four paragraphs, not a single subject is a human being. There was a fucking human inside your car! That person died! And this fact was told by the local cops in Shenmu to a Beijing News reporter that afternoon; it wasn't even mentioned by AITO, this match-seller. The statement didn't even mention how many people were in the car, didn't say how the person died, let alone whether the car doors could still be opened after the collision, nor the vehicle speed at the moment of impact, no braking curve, let alone how long before the crash the assisted driving had disengaged, nor a single word of apology. Even the sign-off was just "AITO User Service"—not a single real human's name, all just corporate tokens. In fact, this is how Huawei perceives the relationship between car owners and Huawei. Seriously, in front of this fucking corporation, Huawei's car owners are all slaves paying out of their own pockets to buy matches. I think this match-seller and the CCP are practically cut from the exact same cloth, because the gaze with which they view all people is identical.

Let's look again at that so-called accident. The CCP's investigation hasn't even concluded on their end, yet this match-seller has already branded the verdict themselves. The absolute kicker is reading the third and fourth paragraphs together. The third paragraph had just said the conclusions are subject to the official investigation. The fourth paragraph immediately turns its own judgment into the standard, declaring that any speculation diverging from its narrative is false information. And this fucking match-seller company acting like the Cyberspace Administration of China isn't something new. That line about "clean and healthy" has practically become a corporate boilerplate for this fucking company Huawei. On June 3 this year, the chassis of an M9 caught fire in Wenling, Zhejiang, and what did AITO User Service say? They said, "Do not believe rumors, do not spread rumors." On July 17 this year, the legal department of their previous brand refuted rumors, with the original words being: "Please do not believe rumors, do not spread rumors, and jointly maintain a clean and upright online environment." On September 5 this year, AITO claimed to have reported to the police in advance, saying rumor-mongers might bear criminal liability. And once again, maintaining a clean and upright cyberspace. In all the automaker statements I reviewed this time, the only one that pasted the entire sentence "jointly maintain a clean and upright online environment" was the announcement from this fucking company. So you see, that is their boilerplate. Hey, so this boilerplate has just been copied and pasted back and forth within a few months. Why does a fucking customer service department of a fucking car company think it is the Cyberspace Administration of China? Hey, I carefully thought about this question. I think in today's public opinion sphere in China, the actual function of the word "rumor" is not to judge true or false, but what is it? It is a pretext for punitive action. Hey, once an account of events is slapped with the label of "false information," platforms proceed with post deletion, accounts face account ban, and the police's admonition has its procedural gateway. Writing "false information" into a corporate statement is not debating with the public; it is handing over a punitive action request to this gang of CCP cops, you know?

Alright, having talked about the rhetoric, let's talk about the data. In the surveillance footage, this car ran a red light and rushed into the intersection at high speed. After crossing the intersection, we can see the brake lights immediately turned on, closely followed by hitting the pole and catching fire. Isn't that steel pole of the traffic light a ready-made speedometer? Because how much energy it can absorb is a fixed physical coefficient determined by the steel material and thickness. Let's first calculate the car's kinetic energy. Hey, it's all middle school physics: kinetic energy equals half the mass times the square of the velocity. Hey, it's all middle school physics stuff, right? So for the range-extended version of the M9, I looked up its weight; it's about 2,650 kilograms. Add a driver, and let's count it as 2.73 tons. Converting 139.1 kilometers per hour into seconds gives 38.64 meters per second. The calculated kinetic energy is 2.04 megajoules, which is equivalent to hoisting this car to the roof of a 25-story building and then dropping it all at once to smash down. If the speed were halved, the kinetic energy would be reduced to just one-quarter. If it were 50 kilometers per hour, it would be a mere 0.26 megajoules.

Now let us look at this pole. The vertical posts for traffic lights are generally 200 to 300 mm square steel pipes; there are pictures online, and the wall thickness ranges from a few millimeters to one centimeter. Let us think about it: what does it equate to when a car hits it? It equals using a lever to forcefully pry the base of the pole, right? How much prying force this base can endure is determined by the strength and wall thickness of the lever. Looking at that pole, the wall thickness is just a square pipe of about 10 mm. Under such circumstances, the front of the vehicle would need to push against it with over 50 tons of force. Just pushing against it is not enough, because if it is merely an instantaneous velocity, the base of the pole would only then begin to bend. Once bent, the pole does not collapse all at once either; it still requires 50 tons of force pushing back continuously all the way. From having this pole standing upright to making it lie flat, the front of the vehicle still needs to squeeze forward by 80 to 90 cm. From start to finish, this pole can absorb at most a total of 0.45 megajoules of energy; it cannot block any more than that. The cross-section of the post at the Shenmu scene is only 200 to 250 mm. Checking its dimensions, it is a small square pipe with a wall thickness of 6 to 8 mm. The vehicle front could bend it with 15 to 30 tons of force. Even if this small square pipe were to completely lie flat, it could only absorb 0.12 to 0.25 megajoules of energy.

Now let us look at the car and the pole together. If this vehicle truly crashed into it at a speed of 139.1 km/h, the energy carried by your vehicle body would be 2.04 megajoules. A large 300 mm pole can absorb at most 0.45 megajoules, leaving 1.6 megajoules remaining in the vehicle body. The speed corresponding to 1.6 megajoules is over 120 km/h. That means after knocking the pole flat, the car would still crush over the pole at a speed of over 120 km/h and continue hurtling forward for dozens of meters; it would be utterly impossible for it to stop right beside the pole and catch fire. Here comes the amusing part: some netizens actually went to the scene to take photos. In the photos, that square post stands upright completely intact—it is not bent, not skewed, let alone broken; only the lower section touching the ground was scorched black by the fire, and a tiny piece merely chipped off the base. Honestly, anything that is called metal, whenever it battles this piece-of-trash car, if it loses, you really have to check whether it is actually metal—it might be plastic. Furthermore, the scorch marks on that post are still there. What does this indicate? It indicates that it is the exact pole from the incident, and it has not been replaced. Based on the proportions in the photos, this post is precisely that small square pipe of 200 to 250 mm. For this type of small square pipe, completely flattening it can only block 0.12 to 0.25 megajoules; yet it did not even bend, which indicates that the energy with which the car hit it was not even 0.2 megajoules. The speed corresponding to 0.2 megajoules is around 50 km/h. Therefore, when this car hit the pole, its speed was at most around 50 km/h, or perhaps even lower. Plus the braking—the brake lights lit up after passing the intersection. Netizens estimated the braking distance frame by frame to be roughly 25 to 40 meters. Calculating based on a dry road surface and assuming full braking at 0.8 g, braking from 139.1 to 50 km/h requires 83 meters—and of course, that is assuming the brakes did not fail; if the road were wet, it would only be longer. Calculating backwards then, if the impact was 50 to 60 km/h, and braking covered 25 to 40 meters, then your entry speed was at most between 87 and 108 km/h. Thus, this standing pole makes it exceedingly clear: at the moment of impact, it was impossible to be traveling at 80 to 90 km/h, nor was there a speed of 139.1 km/h when entering the intersection. That figure was either taken from an earlier moment well before the intersection, or it was an instantaneous reading of the wheel speed at some point. The only thing impossible is the ground speed upon entering the intersection claimed by Match-Seller.

So what is the real black box? It is not this 139.1 reading, but several other numbers that were withheld. According to international EDR standard requirements, a vehicle must record all data for the five seconds preceding a collision, along with numerous accompanying data points—such as vehicle speed at least once every half second, longitudinal acceleration, throttle position, brake status, whether assisted driving was active, and when it disengaged. The cloud backends of these CCP car enterprises record data far more densely and comprehensively than an EDR. In other words, this complete time series is entirely in the hands of Match-Seller. Then the subsequent question becomes: who does this data actually belong to, who has the right to present it, and how to present it? These are all issues. The United States happens to have a precedent on this. In 2015, the US passed the Driver Privacy Act, which expressly stipulates that the data inside the EDR is the property of the vehicle owner, and entities such as car companies, dealerships, and insurance companies cannot unilaterally use it; it can only be retrieved under specific circumstances—such as through a court order, with the consent of the owner, or during investigations by federal safety agencies. In other words, although the person has died, this data still belongs to the deceased's estate and does not constitute PR material for Match-Seller at all. Moreover, even if Match-Seller possesses this data, they cannot simply release it on their own.

Let's look at the 2018 U.S. case. In March 2018, a Tesla Model X crashed into a highway barrier in California, killing the driver on the spot. Tesla did something extremely sneaky. What was it? While the National Transportation Safety Board (NTSB) was still investigating, Tesla preemptively issued a statement saying that Autopilot was engaged at the time and the driver's hands were not detected on the steering wheel six seconds prior to the crash. They added that the vehicle had issued several warnings beforehand. Frankly, that's the exact same tactic as blaming the driver's state and overspeeding in the AITO crash, except Tesla actually mentioned the deceased. The result? The NTSB kicked Tesla out of the investigation completely. The chairman even wrote an open letter stating that releasing incomplete information leads to speculation and incorrect conclusions. For a carmaker to use data from a deceased person during an active investigation to shift blame onto them is an act that gets you immediately disqualified in the United States. Now look back at this Shenmu accident: all the data is held in the hands of these bastards, and no one regulates how they use it. So, out of the entire time-series sequence, AITO cherry-picked the speed at the exact moment the car reached the intersection to disclose, simply because that was likely the absolute peak value on the whole curve—and the most effective point to lock down the narrative that the driver was driving at three times the speed limit. But what about the impact speed? You don't give that instantaneous speed, because that number determines whether the fire onset speed and the escape window were reasonable! You don't provide the braking curve either, because that line shows whether the driver was still trying to control the vehicle in the last two seconds, as well as whether there was an anomaly with the brakes. You don't say how long before the crash the driver-assist system disengaged, because the phrase "manual driving state" only covers the split second of the impact, not the state five seconds, thirty seconds, or three minutes prior. The very existence of international EDRs proves these datasets exist—they just refuse to release them.

While we're at it, let's talk about wheel speed. How does a car measure speed? Usually, a sensor installed on the wheel counts how many revolutions the wheel makes per second, multiplies it by the tire circumference, and calculates the wheel speed. But wheel speed does not equal vehicle speed. True, when the wheels maintain grip on the ground, the faster they spin, the faster the chassis moves—the number is accurate. But once a wheel loses contact with the ground or hits a slippery surface, it no longer has to push a 2.6-ton vehicle body. Once detached from the ground, the motor's power isn't transferred to the vehicle body at all. What is the motor dragging then? Just that remaining tens-of-pounds wheel. The load instantly drops by a hundredfold, and all 5,000 Nm of motor torque goes into spinning a couple of light wheels. In less than a quarter of a second, the wheel speed can spike to that 139.1 km/h number. But the car body wasn't actually moving that fast! Since the car could easily become airborne at the moment of impact, it's highly likely they pulled the wheel speed to spin their narrative. What happens in the opposite case? When you slam on the brakes and lock the wheels, the wheels suddenly stop, but the car body keeps sliding forward. In that case, the speed reported by the sensor is lower than the actual speed of the vehicle body.

In other words, under one condition, the same signal overreports; under another, it underreports. That's why accident appraisals never accept a single-point speed measurement—they require four-wheel speed acceleration curves corroborating with an independent, verifiable speed metric. Yet these matchbox sellers provided none of that. Truth be told, people who buy these cars are really out of their minds. You spend your own money paying for your so-called sentiment and bullshit "far ahead" tech—spend your money how you want, but your family is innocent. Drive a Huawei, end up in the crematory! I think Huawei should just set up a strategic partnership with Fucheng Shares, the top funeral stock in China. After all, with the fucking trash Huawei builds, every major milestone of your life can be wrapped up inside this matchbox. Ah, wait, "far ahead" should actually be in competition with Fucheng Shares. Never mind, pretend I didn't say that. It just reminded me of something funny: in China, every generation has its own Iveco and Gold Cup van. Maybe friends outside the Great Firewall don't know what an Iveco or Gold Cup is—it's a giant van, also known as a land-based hearse. Have you seen the kind used by Hong Kong or Taiwan gangs when they pull up for turf wars?

Yeah, that exact type of van: sliding door flies open, and out pours a bunch of thugs clutching clubs and knives. Later, this van became the standard hearse across northern China. There’s even a二人转 folk tune about it: "Iveco and big Jinbei, hauling corpses first, hauling ashes next." Well, times really have progressed now—Huawei provides a one-stop-shop service: one person sits inside, and everything from driving, to physical exorcism, to flame-throwing immolation happens all inside a single car chassis. Alright, moving on—having talked about the figures, let's talk about the people. Does that garbage post put out by that fucking idiot who sells matches sound anything like a fucking human being? Let me bring up two public statements from other companies for comparison, just so everyone can get a sense.

Back on April 17, 2018, what kind of accident happened on Southwest Airlines Flight 1380 in the US? An engine fan blade shattered a window, resulting in the death of a passenger. Within five hours of the incident, Chairman and CEO Gary Kelly released a video address. What was the very first thing he said? He offered condolences to the victim’s family, emphasizing that the family was the company's most immediate and primary concern. The very next day came a signed apology letter, plus a $5,000 check for every passenger, a $1,000 flight voucher, and not a single word jumping to conclusions about the cause of the accident.

Now let's talk about Toyota. On February 24, 2010, Akio Toyoda testified before the US House of Representatives regarding a recall of over eight million vehicles. That testimony was personally signed by him. What did he say? He said, "I take full responsibility; my name is on every car," admitting that the company's growth had outpaced its capabilities. He sat there enduring four hours of hostile grilling without daring to talk back once. When you hold up a relative benchmark against AITO, AITO's absolute standing immediately shows. Other people's first word is about the person; their first word is about the car. Others sign with the boss; they sign with customer service. Others say "I'm sorry"; they say "Qinglang net-cleansing." What lies between these is not a gap in PR skill, but whether you treat human beings as human beings at all.

So here comes the question: where the fuck does this arrogance come from? Who the fuck gave them this confidence? Arrogance never stems from personality; arrogance is calculated from the ledger. For someone peddling matches to dare talk like this, it must be because across all three fronts—upstream, downstream, and public opinion—they never need to beg anyone for a favor. Let's look at the books: after an AITO is sold, the owner's payment enters Seres's account in full on the very same day. But when Seres buys parts, they don't hand over cash; they issue notes. What kind of notes?

Bank acceptance bills. That means Seres gets a bank to issue a piece of paper stating "payment upon presentation in six months," with the bank providing the guarantee. But the bank's guarantee comes with conditions: Seres has to deposit a sum of margin cash at the bank. Once the suppliers receive this piece of paper, they either front the funds themselves for six months or take it to a bank to get discounted. Discounting means cashing out early, where the bank deducts interest. Now here's where it gets amusing: if we lay Seres's Hong Kong stock annual report and its A-share annual report side by side, at the end of 2025, Seres had notes payable of 38.796 billion, while on that exact same day, its pledged deposits sitting in the bank were 38.908 billion, with the purpose explicitly stated: issuing bank acceptance bills. Just look at that—one to one. By the end of June this year, Seres's notes payable were 38.143 billion, and its restricted deposits were 38.210 billion—still one to one. What does this prove? It proves they didn't rely on the bank's credit at all; they locked the full amount of cash into the margin account and made the suppliers take a piece of paper that could only be redeemed six months later. The suppliers' goods had already turned into the car sold to you, but the car-purchase funds had already turned into their margin deposit.

And what do the suppliers hold in their hands? A slip of paper. Alright, at this point, Seres has already slashed the suppliers once, right? But wait, there's more—the real knife comes next: that margin deposit still earns interest, and this year that interest is even inverted. In a normal year, the bill discount rate should be higher than the deposit interest rate; issuing notes would, at most, amount to tying up the suppliers' money. But this year is different. The rediscount rate on state-owned and joint-stock commercial bank bills plunged at one point to 0.45% this year; in July, the rediscount rate for these bank acceptances plummeted to 0.45%, outright tumbling below 0.5%. At that point, CCP regulators had no choice but to draw 0.5% as a floor that couldn't be breached. Meanwhile, over on the deposit side, Seres's interest income alone last year was 809 million. Divided by its book average of over 60 billion in monetary funds, its overall deposit yield was around 0.2%, and in the first half of this year it was around 0.4%—with one at 0.5% and the other at 0.4%, the deposit rate was nearly a full percentage point higher than the bill rate. This is the bill arbitrage privilege granted by the CCP to the match-seller line.

So, how does this work for Seres? Seres takes the car owners' purchase payments and deposits them as security deposits. At the same time, they issue bills of exchange for the same amount to suppliers. The suppliers then discount these bills, paying an annualized discount rate of 0.5%. This 0.5% is paid by the suppliers; Seres doesn't pay a single cent, they just redeem them at face value upon maturity. Meanwhile, the 1.4% interest on those security deposits belongs entirely to Seres. If we look at Seres and its suppliers together, what are they actually doing? They are essentially borrowing 38.7 billion from the bill market at 0.5% and turning around to deposit it into the bank at 1.4%, netting a profit of nearly one percentage point. Looking at Seres alone, even that 0.5% discount is paid by the suppliers. If calculated at 1.4%, the interest generated by 38.9 billion in security deposits is over 500 million a year; this is leveraging to collect interest. What is the benefit? On the asset side, there are 38.9 billion more in deposits, and on the liability side, there are 38.7 billion more in bills. Both sides rise together, and this 38.7 billion is recorded under accounts payable, not as interest-bearing debt, nor does it enter financial expenses. Therefore, Seres can still write in its interim report today that interest-bearing debt only accounts for 3.2% of total assets.

If you add back those 38.1 billion in bills, it's 34%. So who is subsidizing this one percentage point? It's the banks. Banks are paying Seres that 1.4% deposit interest while holding bills that only yield 0.5%. Why are the banks so fucking stupid to do this? Because bill discounting counts toward credit scale, and security deposits count toward deposit scale. This year, neither companies nor ordinary people dare to borrow money, so what are the banks doing in the mid-to-late stage? They would rather subsidize it than fail to fill these two indicators. So, Seres' interest income last year was 809 million, accounting for 13.6% of the 5.957 billion in net profit attributable to the parent company that year. This year, they lost 1.717 billion on car sales in the first half, but still pocketed 454 million in interest, meaning their interest income this year is even higher than last year's. Looking down, this money is the suppliers' account interest, and looking up, it's because no one in the entire credit market is borrowing money. And this fucking match-selling company takes all the benefits while bearing absolutely no responsibility. Why has this funeral-service-plus-match-selling company been able to make money until today?

Because it has never paid a price. Look, in April 2024, three people died in Yuncheng, not a single car was recalled, no executives came forward, and even the report criticizing that the price of their "far ahead" (遥遥领先) shouldn't be life was censored. Then in November 2024, a Guangzhou car owner produced an appraisal report saying the brakes didn't work, and Seres called the police and filed a lawsuit. In May this year, Seres sued a Zhihu user for defamation, claiming 1 million in damages. At the end of July this year, Luxeed started playing memes, and the match-selling company's legal department filed complaints across the entire web, deleting even posts that didn't mention Huawei. As a result, Luxeed's sales increased sixfold in a week. Then on October 8th, the brake discs of three Stelato V800s were shattered one after another with a single stomp. On October 14th this year, Seres sued another user who posted an AI-faked video, even though that video had less than 3,000 views. The claim was still 1 million. These three 'Jie' brands—Stelato, Luxeed, AITO, whatever—might as well be called the 'Underworld' (Mingjie). They might as well pivot on the spot, use the banner of reviving traditional culture and being 'far ahead' to enter the funeral industry. Anyway, if there are no customers, you can just create them yourselves, right?

Maybe come up with some funeral industry 'set rates' and move them up; that would be environmentally friendly, right? And save land. Now, everyone's disgust for these fucking brands has started to spread. Once the Shenmu incident came out, Seres' stock price, which was already tottering, plummeted from 42 yuan all the way to 29.9 yuan, and along with CATL's trillion-yuan market cap, it plunged 12% in a single day. You say it has nothing to do with this? Fine, the capital market will show you. Also, those fucking things like JAC all hit the limit down. I'm quite shocked that such garbage can even be listed. Let me tell everyone one indicator; this indicator relates to what? It relates to when the A-shares can remove the 'A'. That is, when this market will be playable—it's when all these fucking things die out and are delisted; that's when the A-shares will be playable. As long as these fucking things are here, it clearly shows that it's just a fraudulent market. The day Huawei dies out is the day the Chinese capital market rises. Alright, that's all for today's rant. If you like the show, please bookmark it below. If you want to support me, you can join the membership channel; members get to listen to four episodes per month. Welcome to join the 'Good Guys Gang,' and we'll see you in the next episode.