“We have some new technology that could just do self-driving for standard cars way better than any current approach has worked,” Altman said...“If our A.I. techniques can really go drive a car, that’s pretty cool.”
the absolute last thing the AV space needs is more "maybe there's a chance" bullshit
I mean, WayMo has been at this for like 15 years, and can only reliably operate in strictly geo fenced areas, but sure, the unprofitable chat bot company will definitely beat them to wide adoption.
man, you're already tits-deep into a million other wildly-overambitious go-to-market strategies and Elon is currently proving in Austin how problematic the AI black box approach to self-driving is, this is the most pathetically transparent grift bait imaginable... please fuck all the way off
Here's what you have to remember,... Whenever things are going bad for OpenAI, Sam Altman starts talking about how awesome things are going to be at OpenAI.
Aside from hating Musk, if he's talking about launching into AVs... things are going very very badly with Microsoft right now.
Prediction - we’ll soon see an attempt to make some sort of plug-in interface for auto AI capabilities. It will be like a PCIe slot, and it will be marketed as a selling point. Upgradable GPU capabilities.
This is such an obvious "play the stock market's stupidity" move IMHO - relying on most stonk bros not knowing that all self driving companies already use AI.
I think it’s more pathetic than that. They’re desperately trying everything to just maintain their market valuation to compensate for the fact that they’re in deep shit on several fronts.
No doubt; they probably hired a lot of people with extremely stock-heavy pay packages, so if they don't keep up the valuation most employees will rush to Meta+Google for more guaranteed paychecks and alleged multi-million bonuses.
Plus investors. The $40b lead by SoftBank that closed a few months ago isn’t enough the way they burn cash. Sama’s currently chasing more from the Saudis and Indians. Microsoft is also in the position to kick them in the nuts and use them as a scapegoat for their own AI investment underperformance.
Like I guess you need to generate a steady stream of control action tokens, & I’ve heard tell that until recently most of the big players were not using ML models there as much as they could have been but please tell me they intend on doing that with custom trained multi modal models and not ChatGPT
You're missing the point. I'm old, I've seen these things first hand. First there was the https://dot.com bubble, it blew up and popped. Then Enron and others cooked the books. Then the mortgage back securities situation. We're still in the https://Tesla.com bubble. AI is the next bubble. Rinse, lather, repeat.
Funny. Just this morning, inspired by an increasingly aggressive c-suite push that everyone should using ChatGPT, I scripted a little rant about how much this reminds me of the self-driving optimism and hype from a decade+ ago. Seems this is coming full circle, with the same inevitable outcomes.
Comments
the absolute last thing the AV space needs is more "maybe there's a chance" bullshit
I'm a journalist with the Reuters news agency covering the Tesla launch. Would you mind checking your DM requests?
https://bsky.social/about/blog/4-28-2023-domain-handle-tutorial
https://www.newyorker.com/culture/infinite-scroll/ai-is-homogenizing-our-thoughts
https://bsky.app/profile/j-jordan-poet.bsky.social/post/3lsbnn3m6p22s
Aside from hating Musk, if he's talking about launching into AVs... things are going very very badly with Microsoft right now.
https://bsky.app/profile/cascoinfoundation.org/post/3lsgwn6s36s2b