|
On November 22 2024 10:41 Magic Powers wrote:Show nested quote +On November 22 2024 10:24 ETisME wrote:On November 22 2024 09:18 Magic Powers wrote:On November 22 2024 08:33 BlackJack wrote:On November 22 2024 07:54 Magic Powers wrote:On November 22 2024 07:08 BlackJack wrote:On November 21 2024 21:15 Magic Powers wrote:On November 21 2024 20:38 BlackJack wrote:On November 21 2024 19:56 Magic Powers wrote:On November 21 2024 14:52 Turbovolver wrote:
[quote]
Not really a big deal. Tesla was higher by being 5.6 (fatalities per billion miles) as compared to 5.5 for Kia. Autonomous vs non-autonomous, my friend. "Not a big deal"? Reality says otherwise. Hyundai has 3.9 Good lord... The article showing the fatality rate says it used data from 2017 to 2022. Tesla didn't even have a wide release for FSD until November of 2022 and even then only 2% of their drivers paid for it after the free trial. Not only have you provided zero evidence that Tesla's slightly higher fatality rate is related to autonomous driving but the timeline makes it an implausibility. You're a fedora-wearing conspiracy theorist with an Elon Musk hate-boner trying to connect dots that don't even make sense. But at least we found out who is the audience for the fearmongering news stories. All autonomous cars have a higher accident rate than regular cars. More fatalities by Tesla are completely expected. The finding is that Tesla is no better than its competition and certainly not better than regular cars except environmentally. *citation needed* Citation for what? The first sentence? Here you go. https://www.nature.com/articles/s41467-024-48526-4 The analysis suggests that accidents of vehicles equipped with Advanced Driving Systems generally have a lower chance of occurring than Human-Driven Vehicles in most of the similar accident scenarios. However, accidents involving Advanced Driving Systems occur more frequently than Human-Driven Vehicle accidents under dawn/dusk or turning conditions Based on the model estimation results, it can be concluded that ADS in general are safer than Human-Driven Vehicles in most accident scenarios for their object detection and avoidance, precision control, and better decision-making. So your evidence for your claim "All autonomous cars have a higher accident rate than regular cars" is a study that concludes that autonomous cars are safer than human driven vehicles in "most" scenarios but are less safe in dusk/dawn conditions and when turning. You're not even a stone's throw away from proving your claim. In fact you're closer to proving the opposite. I'm not even going to make an effort to explain why your reasoning is wrong. We have history regarding scientific papers, you tend to heavily misread them and then never acknowledge your mistake. I'll just let it stand cause I'd be wasting my time. your own paper doesn't even reference tesla having higher accident rate than other ADS. And it even states: On one side, numerous studies support the view that AVs are generally safer than HDVs.....On the other hand, some research challenges this view, suggesting that the safety of AVs may not always exceed that of HDVs How did you interpret this sentence? I refer to the other links I posted earlier. You can ignore reality if you like. Btw I fully support autonomous cars, but the technology just isn't there yet. They may eventually be good enough to replace regular cars.
"The authors of the study, published in November, noted that Teslas with their advanced driving technology were safe cars but that did not mean their owners were safe drivers.
“Most of these vehicles received excellent safety ratings, performing well in crash tests at the IIHS and NHTSA, so it’s not a vehicle design issue,” said the company's executive analyst Karl Brauer.
“The models on this list likely reflect a combination of driver behaviour and driving conditions, leading to increased crashes and fatalities.”"
Do you mean the website with this quote?
If we are to follow research data etcetc, then pretty sure your own link highlight research show consistent result for AV to be on par if not safer in general:
"On one side, numerous studies support the view that AVs are generally safer than HDVs.....On the other hand, some research challenges this view, suggesting that the safety of AVs may not always exceed that of HDVs"
|
On November 22 2024 10:53 ETisME wrote:Show nested quote +On November 22 2024 10:41 Magic Powers wrote:On November 22 2024 10:24 ETisME wrote:On November 22 2024 09:18 Magic Powers wrote:On November 22 2024 08:33 BlackJack wrote:On November 22 2024 07:54 Magic Powers wrote:On November 22 2024 07:08 BlackJack wrote:On November 21 2024 21:15 Magic Powers wrote:On November 21 2024 20:38 BlackJack wrote:On November 21 2024 19:56 Magic Powers wrote:
[quote]
Autonomous vs non-autonomous, my friend. "Not a big deal"? Reality says otherwise. Hyundai has 3.9 Good lord... The article showing the fatality rate says it used data from 2017 to 2022. Tesla didn't even have a wide release for FSD until November of 2022 and even then only 2% of their drivers paid for it after the free trial. Not only have you provided zero evidence that Tesla's slightly higher fatality rate is related to autonomous driving but the timeline makes it an implausibility. You're a fedora-wearing conspiracy theorist with an Elon Musk hate-boner trying to connect dots that don't even make sense. But at least we found out who is the audience for the fearmongering news stories. All autonomous cars have a higher accident rate than regular cars. More fatalities by Tesla are completely expected. The finding is that Tesla is no better than its competition and certainly not better than regular cars except environmentally. *citation needed* Citation for what? The first sentence? Here you go. https://www.nature.com/articles/s41467-024-48526-4 The analysis suggests that accidents of vehicles equipped with Advanced Driving Systems generally have a lower chance of occurring than Human-Driven Vehicles in most of the similar accident scenarios. However, accidents involving Advanced Driving Systems occur more frequently than Human-Driven Vehicle accidents under dawn/dusk or turning conditions Based on the model estimation results, it can be concluded that ADS in general are safer than Human-Driven Vehicles in most accident scenarios for their object detection and avoidance, precision control, and better decision-making. So your evidence for your claim "All autonomous cars have a higher accident rate than regular cars" is a study that concludes that autonomous cars are safer than human driven vehicles in "most" scenarios but are less safe in dusk/dawn conditions and when turning. You're not even a stone's throw away from proving your claim. In fact you're closer to proving the opposite. I'm not even going to make an effort to explain why your reasoning is wrong. We have history regarding scientific papers, you tend to heavily misread them and then never acknowledge your mistake. I'll just let it stand cause I'd be wasting my time. your own paper doesn't even reference tesla having higher accident rate than other ADS. And it even states: On one side, numerous studies support the view that AVs are generally safer than HDVs.....On the other hand, some research challenges this view, suggesting that the safety of AVs may not always exceed that of HDVs How did you interpret this sentence? I refer to the other links I posted earlier. You can ignore reality if you like. Btw I fully support autonomous cars, but the technology just isn't there yet. They may eventually be good enough to replace regular cars. "The authors of the study, published in November, noted that Teslas with their advanced driving technology were safe cars but that did not mean their owners were safe drivers. “Most of these vehicles received excellent safety ratings, performing well in crash tests at the IIHS and NHTSA, so it’s not a vehicle design issue,” said the company's executive analyst Karl Brauer. “The models on this list likely reflect a combination of driver behaviour and driving conditions, leading to increased crashes and fatalities.”" Do you mean the website with this quote? If we are to follow research data etcetc, then pretty sure your own link highlight research show consistent result for AV to be on par if not safer in general: "On one side, numerous studies support the view that AVs are generally safer than HDVs.....On the other hand, some research challenges this view, suggesting that the safety of AVs may not always exceed that of HDVs"
What are we in disagreement about? Data may suggest either conclusion, so what are we arguing about? You can say that autonomous cars are doing fine, you cannot argue they're better. For now. Tesla has been reviewed as well, it performs worse than some regular brands.
|
On November 22 2024 07:54 Magic Powers wrote:Show nested quote +On November 22 2024 07:08 BlackJack wrote:On November 21 2024 21:15 Magic Powers wrote:On November 21 2024 20:38 BlackJack wrote:On November 21 2024 19:56 Magic Powers wrote:On November 21 2024 14:52 Turbovolver wrote:Not really a big deal. Tesla was higher by being 5.6 (fatalities per billion miles) as compared to 5.5 for Kia. Autonomous vs non-autonomous, my friend. "Not a big deal"? Reality says otherwise. Hyundai has 3.9 Good lord... The article showing the fatality rate says it used data from 2017 to 2022. Tesla didn't even have a wide release for FSD until November of 2022 and even then only 2% of their drivers paid for it after the free trial. Not only have you provided zero evidence that Tesla's slightly higher fatality rate is related to autonomous driving but the timeline makes it an implausibility. You're a fedora-wearing conspiracy theorist with an Elon Musk hate-boner trying to connect dots that don't even make sense. But at least we found out who is the audience for the fearmongering news stories. All autonomous cars have a higher accident rate than regular cars. More fatalities by Tesla are completely expected. The finding is that Tesla is no better than its competition and certainly not better than regular cars except environmentally. *citation needed* Citation for what? The first sentence? Here you go. https://www.nature.com/articles/s41467-024-48526-4
I don't want to go into this too deeply, as the main message from that paper appears to be that we don't know, the data isn't there yet, and anyway to do any kind of statistics analysis we need to combine vastly different levels of autonomy and generations of the technology to say anything at all.
That said, there is a very obvious bias here, which is that self-driving cars switch the self-driving part off in challenging conditions. Snow on the road? Noping out. Hard rain? Noping out. Busy inner city? Noping out.
It's fairly obvious to me that AI is going to.reduce incidents on highways where the 2 main reasons for accidents are driving top fast, and getting distracted/falling asleep. That is exactly the sort of thing AI is going to be good at. Predicting that a child outside of vision might run into the street when a ball rolls into said street? Not something AI is going to be good at (yet).
Further comment is that Elon Musk intentionally and very personally disallowed the use of any non-vision tools on Tesla's AI platform. It seems like an extremely dumb self-pwn and I've talked to engineers who are 100% convinced that Tesla would still be leading the field in self-driving if they hadn't changed course. Instead other providers have caught up and even overtaken Tesla in this domain. Vision is notoriously hard, so why limit yourself if you can equip the thing with LIDAR and a host of other sensor technology that is better at detecting things both further away and very near to the car. I really didn't understand the choice at the time and in hindsight I have seen nothing to change my mind.
|
On November 22 2024 11:38 Magic Powers wrote:Show nested quote +On November 22 2024 10:53 ETisME wrote:On November 22 2024 10:41 Magic Powers wrote:On November 22 2024 10:24 ETisME wrote:On November 22 2024 09:18 Magic Powers wrote:On November 22 2024 08:33 BlackJack wrote:On November 22 2024 07:54 Magic Powers wrote:On November 22 2024 07:08 BlackJack wrote:On November 21 2024 21:15 Magic Powers wrote:On November 21 2024 20:38 BlackJack wrote:
[quote]
Good lord... The article showing the fatality rate says it used data from 2017 to 2022. Tesla didn't even have a wide release for FSD until November of 2022 and even then only 2% of their drivers paid for it after the free trial. Not only have you provided zero evidence that Tesla's slightly higher fatality rate is related to autonomous driving but the timeline makes it an implausibility. You're a fedora-wearing conspiracy theorist with an Elon Musk hate-boner trying to connect dots that don't even make sense. But at least we found out who is the audience for the fearmongering news stories. All autonomous cars have a higher accident rate than regular cars. More fatalities by Tesla are completely expected. The finding is that Tesla is no better than its competition and certainly not better than regular cars except environmentally. *citation needed* Citation for what? The first sentence? Here you go. https://www.nature.com/articles/s41467-024-48526-4 The analysis suggests that accidents of vehicles equipped with Advanced Driving Systems generally have a lower chance of occurring than Human-Driven Vehicles in most of the similar accident scenarios. However, accidents involving Advanced Driving Systems occur more frequently than Human-Driven Vehicle accidents under dawn/dusk or turning conditions Based on the model estimation results, it can be concluded that ADS in general are safer than Human-Driven Vehicles in most accident scenarios for their object detection and avoidance, precision control, and better decision-making. So your evidence for your claim "All autonomous cars have a higher accident rate than regular cars" is a study that concludes that autonomous cars are safer than human driven vehicles in "most" scenarios but are less safe in dusk/dawn conditions and when turning. You're not even a stone's throw away from proving your claim. In fact you're closer to proving the opposite. I'm not even going to make an effort to explain why your reasoning is wrong. We have history regarding scientific papers, you tend to heavily misread them and then never acknowledge your mistake. I'll just let it stand cause I'd be wasting my time. your own paper doesn't even reference tesla having higher accident rate than other ADS. And it even states: On one side, numerous studies support the view that AVs are generally safer than HDVs.....On the other hand, some research challenges this view, suggesting that the safety of AVs may not always exceed that of HDVs How did you interpret this sentence? I refer to the other links I posted earlier. You can ignore reality if you like. Btw I fully support autonomous cars, but the technology just isn't there yet. They may eventually be good enough to replace regular cars. "The authors of the study, published in November, noted that Teslas with their advanced driving technology were safe cars but that did not mean their owners were safe drivers. “Most of these vehicles received excellent safety ratings, performing well in crash tests at the IIHS and NHTSA, so it’s not a vehicle design issue,” said the company's executive analyst Karl Brauer. “The models on this list likely reflect a combination of driver behaviour and driving conditions, leading to increased crashes and fatalities.”" Do you mean the website with this quote? If we are to follow research data etcetc, then pretty sure your own link highlight research show consistent result for AV to be on par if not safer in general: "On one side, numerous studies support the view that AVs are generally safer than HDVs.....On the other hand, some research challenges this view, suggesting that the safety of AVs may not always exceed that of HDVs" What are we in disagreement about? Data may suggest either conclusion, so what are we arguing about? You can say that autonomous cars are doing fine, you cannot argue they're better. For now. Tesla has been reviewed as well, it performs worse than some regular brands.
You're the one making the positive claim that "All autonomous cars have a higher accident rate than regular cars." The disagreement is whether you've made that case, especially when your only evidence is a study that conclues accidents have a lower chance of occuring in ADS than in human driven vehicles in most scenarios.
https://arxiv.org/pdf/2312.12675
This paper analyzes the safety of Waymo's (parent company Google) autonomous cars. They found human driven vehicles were 5x as likely to be in an injury-reported crash than their autonomous vehicles. I guess in this reality that we are denying "5x more" actually means "less."
|
|
|
|
|
|
EPT
