IMO the benchmark to pass will be to show it has better safety than the average sober, non-speeding, non-phone-using, seatbelt-wearing driver.
IDK how that could even feasibly be shown but that's when people will actually start going all-in on it. Otherwise I will simply meet those criteria myself for free.
Bordering on zero human drivers fit that profile. The competition to AVs isn’t a mythical race of flawless humans. It’s the drunk, speeding, distracted idiots who currently operate motor vehicles (e.g., us).
The trick is, LLMs are only as good as they are at writing code because of all the publicly available source code, tutorials, blog posts and Q&A it can slurp up.
Well obviously cell phones have very powerful tx/rx so my question would be "what is stopping us from using this for p2p", I assume the answer is we don't have software access to the radio, and I assume that's for regulatory reasons but idk
That's not actually true: cell phone rx/tx power is quite low. We can get away with that because all they need to do is get to the nearest tower, which has a ton of power, sensitive antennas, and is very tall. Amateur radios have far more power available to them, but any "p2p" (i.e. simplex in amateur radio) runs into normal RF issues, like obstacles and interference. If you used the existing radios in cell phones to communicate directly with other cell phones, you wouldn't get very far. Even amateur radios, with all their power, use repeaters to the same effect as cell towers.
IDK how that could even feasibly be shown but that's when people will actually start going all-in on it. Otherwise I will simply meet those criteria myself for free.
reply