So I'm assuming you're fine with regular drivers using basic lane keep systems from other companies, which honestly doesn't even work well, even in the latest cars. (there's a reason Comma.ai exists) At least people who are using FSD are enthusiasts and understand the tech. You have some people using lane keep with adaptive cruise control and think the car is "self driving". That's dangerous.
The tools are going to ~zero (~ 5 years). The open source LLM's are here. No one can put them back or take them down. No internet, no problem. I don't see a long term future in frontier llm companies.
What I don't get is, how are these free LLMs getting funded? Who is paying $20-100 million to create an open weights LLM? Long term why would they keep doing it?
I see what you're saying, but it doesn't matter that much in the long run. If everything stopped right now, the state-of-the-art open source models can still solve a lot of problems. They may never solve coding, per se, but they're good enough.
Tesla AP is a basic ADAS system like Toyota Safety Sense. So using FSD and AP in the same sentence basically reads like Microsoft Windows 3.1 and Microsoft Windows 11.
99% of my driving is FSD. Yea it took a while but with FSD 14 it's finally here. No other consumer self-driving is close. If you saw what this thing could do you would never not use it. It's like driving without a seatbelt now. You would never not want to be driving without it.
I had a former coworker who loved letting the FSD drive him places. He used to love driving people for lunch, ignoring all the crazy red flags like the car's inability to consistently detect other vehicles on the road around it or stay in the (clearly marked) lanes.
Then one day on the way home from work FSD nearly killed him.
You're not agreeing with reality because you haven't used FSD 14. You're making assumptions on headlines and older versions. And honestly, 95% of the time, people are making assumptions about FSD when, in reality, they're thinking about Autopilot.
"You'll be in an autonomous car full-time in 10 years, no matter what"
Hmmm... maybe. It's possible that in 10 years I'll still be driving the car I drive today, although not that likely. Pretty sure my next car will be electric. Autonomous I'm not so sure about. We'll see, I guess.
I don't entirely get the appeal of a self-driving car. What am I going to do while the car drives itself? I can't read in a moving vehicle. On a long trip, the most likely thing I would do is fall asleep and that does not seem like a great idea.
Unless it’s icy or rainy or you’re on a dirty road and the cameras can’t see anything. Enjoy getting out to wash your camera lenses off every quarter mile.
That's an interesting point. One, I wish I made haha. This article is for people who are "into" this stuff (tech). Who live and breathe it. Who've been doing it as a kid and just getting into agentic coding.
You didn't read the article. The TLDR is: treat it like learning anything else (Like learning a language). Go into it full force with good faith, and you'll have a better outcome.
I think I can live off just the fire ants in my yard. Not to mention all the neighbors. Even with constantly baiting them, it's hard not have at least 1 hill active at all times during the summer, as the neighbors are dumb and just spray. So they just end up moving.
People are so worried about the 1% of driving. 99% of the driving is in town, and in that case, it's super convenient. Yeah, you might have to stop an extra time on a road trip, so what? The car is also driving itself 99% of the trip. Plus, the kids have to pee.
If you have to drive 250 miles every day for work, or you don't have kids and can do 3 hour legs on trips without stopping, then get a gas car.
People think you can just slap lidar on and poof, self-driving is solved. There still has to be a software / ML stack. You still have to know what you're doing. You still need a lot of data.
reply