mikejb

7 ms·

1.
by mikejb 5y ago
This is from from November 2021, but I'm still highlighting it because it is just terrifying (Correct time, though the video later on also exhibits inabilities of the system): https://youtu.be/9wRRClg_aM8?t=113
2.
by mikejb 5y ago
I think it's not just that. When Tesla started to with their FSD journey, they had to determine what sensors they can add to the car. Lidars back then were way more expensive than they are now, and it wouldn't have been feasible t
3.
by mikejb 5y ago
Another country that comes to mind is India - but I've never been enough of a Bollywood fan to be able to make any assessment on this topic. But it's a very effective tool to shape public opinion. I'm fairly certain without t
4.
by mikejb 5y ago
As much as it would be a solution to Tesla's problem, it would impact Tesla's overall plan and strategy. Automatically disengaging when emergency vehicles are detected is a huge barrier in their selling strategy ("It can driv
5.
by mikejb 5y ago
> I'm not seeing this with the current situation It's what Waymo's is currently doing. So far, you only can get automated Taxis (in some regions) - you cannot buy an automated vehicle yet. As I see we do agree that they
6.
by mikejb 5y ago
I think this also translates to self-driving cars. I actually think automated cars have an edge there, because of 2 reasons: 1. Improved visibility: A human driver sits in the driver seat, and looks at the outside world from inside the car
7.
by mikejb 5y ago
I think you're confusing a Tesla with a self-driving car.
8.
by mikejb 5y ago
Ok, I think we're having 2 separate conversations: a) What's the impact of self-driving cars at what point of distribution and b) What's the distribution of self-driving cars going to be. Regarding the impact: modern vehicles
9.
by mikejb 5y ago
Auotmated drivers don't make the human driver safer - they (should) make traffic safer by reducing the number of human drivers that would cause accidents. I don't think we're there yet, but I also don't think this is cen
10.
by mikejb 5y ago
I think Lvl3 is one of the most dangerous levels a car can be at, based on the experience Waymo (back then Chauffeur) made: Humans are even worse at monitoring than they are at driving. It's incredibly boring. Their test-drivers (who
11.
by mikejb 5y ago
To be fair: Self-driving cars that can drive safer than some drivers already exist. Simply due to the nature how unsafe some humans drive. (Some of these drivers are elderly, some of these drivers are racers who think everyone slower than
12.
by mikejb 5y ago
I have my doubts about V2V and I2V communications. They'd simplify the task for automated systems, but come with a rat tail of problems. I2V is incredibly hard to scale, and is a duplicated implementation of a pre-existing system (road
13.
by mikejb 6y ago
That particular attack had been fixed, but I remember people disabling those warnings with oranges: https://www.youtube.com/watch?v=TYZrehVQouc It's an arms race between Tesla being legally on the safe side, and driver
14.
by mikejb 6y ago
> you always have plenty of warning before you need to take over This only has to be false once. I'd also like to highlight that sometimes AP has frightening regressions. One case was labeled "barrier lust", where - on a d
15.
by mikejb 6y ago
Right, because who would digitalize the entire world and take pictures of every house and store front - that'd be insane!
16.
by mikejb 6y ago
This is a curious point: How much data does Tesla actually collect? I can imagine users not being too happy if their car clogs the upload for hours every time you park it in your garage?
17.
by mikejb 6y ago
To be fair, Waymo (back then Chauffeur) have solid experience with Tesla's approach. They started with the same strategy of gradual improvements. They found out (and shard) the same issues Tesla is running into in the past years: Users
18.
by mikejb 6y ago
Whilst Lidar and cameras may have similar wavelengths, they have very different information outputs. Cameras give you a 2D colored image from which you essentially guesstimate the distance of objects. Lidar gives you a 2D depth map with muc
19.
by mikejb 6y ago
Your 'experience' explains a lot about your opinion. Have a good day, stay safe.
20.
by mikejb 6y ago
So you say you suspect they have an ongoing investigation, and that would be "circumstances where every half competent organization would decline comment" What substantiates this conclusion? They can have an investigation ongoing,
21.
by mikejb 6y ago
> And even if the ethical obligations were identical, Gebru’s violation toward Google wouldn’t excuse Google’s toward Mitchell. What about Mitchell's violation towards Google?
22.
by mikejb 6y ago
Just to get this straight: The researcher who tweeted out the name of an employee who's email has been blocked, and throwing out theories about crackdowns and firings - that's all good. Explaining the email account has been blocke
23.
by mikejb 6y ago
> The normal response (for a variety of good reasons, including legal and ethical ones) from a company on a matter like this when asked by the press when they haven’t completed an internal investigation would be to decline comment on per
24.
by mikejb 6y ago
The researcher who's departure sparked a lot of controversy tweeted the name of the employee in question, with some opinion attached. Axios reached out to Google who then made the statement saying security systems detected mass-forward
25.
by mikejb 6y ago
Not every leak is a whistleblow. It depends on the content of the files. That verdict still needs to be made (depending on the content of the leaked files).
26.
by mikejb 6y ago
The (1) case is a bit inaccurate/misleading. From what I can gather from the article: - Gebru tweeted the name of the employee [1] - Axios then reached out to Google, who then made the following statement: > Our security systems aut
27.
by mikejb 6y ago
I couldn't find the direct quote, so they may have adjusted the sentence you're referring to. It's clearer now: > If you're inactive in one or more of these services for two years (24 months), Google may delete the co
28.
by mikejb 6y ago
Lidar-based systems are pretty good at detecting stationary objects. Humans on the other hand crash into traffic jams pretty regularly. So in that situation I'd prefer that to be an automated truck.
29.
by mikejb 6y ago
I'd say the breakthrough term was not the ideal word choice - it's more that technological advances passed a threshold: For VR, a lot was advances in (affordable) processing power (latency causes significant issues in VR), and scr
30.
by mikejb 6y ago
I named these 2 companies because the parent referenced them in their argument. But I agree that they are the ones that are the ahead in their individual approaches, though I think Tesla is not really that near to autonomous driving. They h

More ›