Honest assertions about the AI bubble
- Thread starter firegiver
- Start date
Of course. My point remains. From the mundane to the super intricate. Robots, AI, etc are the future. Amazon says they want to eliminate 1 million jobs out loud. Every major company will move this direction from a profit standpoint. That’s when it gets tricky imo.
right and investors are spendign their capital on the speculation of that return. The PE for Nvidia is 50 right now.
My point is: theres a TON of investing going on, but there is no profit yet. Thats the point im making.
Waymo is lapping the competition today. Anyone who is following along knows they can be put out of business overnight. Once Robotaxi is released, it's over for Waymo. They can't compete on cost.
Grok says the production line to produce 1 million Optimus a year will be finished by the end of 2026. Current websites show Tesla is hiring for the Optimus production line. While there may be delays, saying it is a decade away is too pessimistic in my eyes.
Grok:
Tesla's humanoid robot, Optimus, is not yet in mass production as of October 2025. During the company's Q3 2025 earnings call on October 23, CEO Elon Musk announced that the Optimus Version 3 prototype will be unveiled in Q1 2026 (early next year). Mass production is targeted to begin late in 2026, with plans to install a dedicated production line capable of 1 million units per year by the end of that year. This timeline reflects a delay from earlier 2025 goals of producing 5,000–10,000 units, which were abandoned in October 2025 due to unresolved design challenges in the robot's hands and arms.
Of the Mag 7, NVDA trades the cheapest on a FWD PE ratio.
Grok:
Yes, NVIDIA (NVDA) is highly profitable. For its fiscal year 2025 (ended January 26, 2025), the company reported net income of $72.88 billion, a 145% increase from the prior year.NVIDIA generated $60.853 billion in free cash flow during fiscal year 2025, up 125% from fiscal year 2024.
Grok:
Yes, NVIDIA (NVDA) is highly profitable. For its fiscal year 2025 (ended January 26, 2025), the company reported net income of $72.88 billion, a 145% increase from the prior year.NVIDIA generated $60.853 billion in free cash flow during fiscal year 2025, up 125% from fiscal year 2024.
Wasn’t Tesla that way for years? I think that’s somewhat expected in up and coming markets, that’s part of the chase. I tend to agree with some of your points but many of these companies are very undervalued when looking at PEG which is also a great indicator. IREN is way undervalued and they would fit your description of speculative. I partner what I see in the market with what I see in real life. People have never been more dependent on AI type things ever, it’s not slowing down imo. I think the next big market will be blockchain tokenization in a sense that each person will have a token that’s unique and independent to them. It will be how we know what’s real and what’s AI. There’s zero profit in that today but it’s inevitable wouldn’t you agree?
I hope it's not inevitable. Welcome to social credit scores if it does. You are describing a CBDC.
I'd much prefer BTC.
I don’t like it but it’s going to be where we are headed. AI will become indiscernible from reality. There will need to be a way to confirm each person. Leaders will have tokens that will ensure it’s them speaking and not AI, etc. I worked at a place that did this. It’s crazy.
I think you will see resistance from the people. We don't want social credit scores here. It's total bullsh!t.
Don't say the right thing about the president? Well, we will cut off your ability to spend money.
I hope it never takes hold. AI is already indecipherable from real life, but we don't have any problems yet.
Well, the head of robotics left a few months ago which isn’t a good sign. Musk is also more optimistic than dabo, so you can always discount his projections significantly.
From talking with friends who are deploying a few robotics deployments in logistics facilities, they are really struggling to incorporate robotics into workflows. Perhaps they get orders that are non committal, but the market is just not ready to adopt at the rate that musk wants to sell these. Nor is their tech anywhere near ready to handle the tasks to make this a viable alternative to human labor.
Musk has pulled off some amazing things in his career, but this effort is very different from rolling out EVs, offering satellite launches at a significantly lower cost, and offering WiFi in rural areas. The market didn’t have to do any work to change over to those products, whereas Optimus will require massive rearchitecting of operations.
Waymo is stewed. I'm telling you the update my car got this morning is absolutely nuts. I just don't think people understand how good it is because most people don't drive teslas. Unless you have experienced it you have no idea how good it is. It is ready for primetime right now, just as Morgan Stanley analyst Adam Jonas says in his new Tesla analysis.
Morgan Stanley's Adam Jonas in new $TSLA note: "I'm callin' it. Autonomous cars are solved. Do I mean six or seven 9's to the right of the decimal? No. Perfection? Never. But enough to pull the safety driver at scale in major metros." "From our discussions with
@Tesla
, it appears the only thing preventing Tesla from pulling the driver is its own abundance of caution. Our understanding is that there is no other explicit regulatory approval required to pull drivers in Texas. A passive optical-only AV system would seriously challenge the conventional thinking of many in the robotaxi community. Tesla truly believes that radar and LIDAR do not make its cars any safer. It is our longstanding opinion that if you solve autonomy for cars – you solve autonomy for many other form factors of AI-enabled robotics (aviation, marine, weapons, etc). Tesla will have around 8 million vehicles in the global 'parc' by year-end. 12% FSD penetration (paying monthly subs and upfront purchasers) implies 1mm active paying users. If we were to apply the $99/month subscription fee to all users (including upfront payers for comparative equivalency) implies approx $300mm of quarterly FSD revenue or $1.2bn annualized FSD revenue at a year-end exit rate. Assuming an 85% margin would imply approx $1bn of EBIT... accounting for 15 to 20% of total company EBIT. Highly profitable, highly reoccurring."
Here’s the dirty little secret. All these analysts are saying nice things because they want Elon’s dealflow whether it’s financing Xai, or taking space x public, or working on the next company secondary tender.
Anyone with half a brain knows that AVs are very different than other robotics. Teslas neural net is trained on road traffic and would require building a completely different neural net for all those use cases. Tesla also has a crash rate that is about equal to human drivers - with a human in the loop! Waymo’s crash rate is significantly lower (~75%) than human drivers.
Thats not to say it isn’t amazing tech - it is!!! But it’s not anywhere near Waymo and not anywhere near ready for robotaxis.
OpenAI and Anthropic will generate close to $60 billion in revenue next year.
Require building a completely different neural net - wrong
Crash rate equal to human drivers. - wrong (10x better)
I don't know where you are getting your info from. And Jonas is one of the more respected analysts out there.
Robotaxi is ready for primetime right now. I use it every day and it's ready. I'd feel comfortable getting in one, certainly a much better than my 16 year old daughter who just got her license last week.
Not sure where you are getting your info from Rasta
Grok:
Overall, FSD/Autopilot crashes ~9x less frequently than human-driven vehicles per mile, substantiated by billions of real-world miles. As FSD scales (e.g., robotaxi pilots), expect further divergence.
Compare FSD to Waymo. Grok:
In summary, Waymo is not measurably safer overall—FSD edges it out 2-3x on serious crash rates, though Waymo leads in zero-fatality unsupervised scaling. Direct head-to-heads (e.g., Business Insider tests) favor Waymo for reliability in complex scenarios, but aggregate data tilts toward FSD for raw efficiency. Expect convergence as FSD unsupervised rolls out in 2026.
Grok:
Overall, FSD/Autopilot crashes ~9x less frequently than human-driven vehicles per mile, substantiated by billions of real-world miles. As FSD scales (e.g., robotaxi pilots), expect further divergence.
Compare FSD to Waymo. Grok:
In summary, Waymo is not measurably safer overall—FSD edges it out 2-3x on serious crash rates, though Waymo leads in zero-fatality unsupervised scaling. Direct head-to-heads (e.g., Business Insider tests) favor Waymo for reliability in complex scenarios, but aggregate data tilts toward FSD for raw efficiency. Expect convergence as FSD unsupervised rolls out in 2026.
lol, you can’t just respond with “wrong”. Flying a drone or a sub require completely different neural nets. They aren’t dodging other cars, humans crossing streets, etc.
Here is a recent post showing that what Tesla is reporting is manipulated and the numbers are getting worse.
Tesla's Autopilot safety data is getting worse
Tesla has released its latest Autopilot safety report, and the limitations are still presented misleadingly; however, one clear thing is...
electrek.co
I’ll try to find the report that showed autopilot was on par with human drivers. But there is a reason Tesla isn’t transparent with this stuff.
It also is heavily weighted towards highway miles, which have much lower incidents. Almost all of Waymo is dense urban miles, making that even more impressive
You are turning into the “grok is it true meme”.
Yes, of course they run on neural nets. That is the primary ML model architecture for computer vision. And I’m sure they are running on the same cloud as well. Not sure what is your point. They require different models. I kinda know this stuff.
I like all the misspellings and the assertions without any actual arguments. Just insults hilarious.
In other news: https://globalnews.ca/news/11498213/openai-chatgpt-for-profit-business-structure/
I’m trying to answer your question. The guy is a retired Goldman Sachs m&a banker, he knows his ****. Best follow on X and the $5 subscription to his info is the best value on the internet.
I was trying to say don’t take it from me, take it from someone uniquely qualified. You started a thread asking questions so I figured you would want the info.
*The misspelling are intentional so as to not get deboosted by the X algorithm.
You are a bit sensitive.
Well, that is either an incredibly stupid thing for musk to say. Or he is just blatantly lying again. ChatGPT went from 0-800 million users in two years. Tesla has sold 8 million cars to date. Yes, if and when they figure out robotaxis, they could flip the switch and have 8 million robotaxis in a matter of minutes. But there are dozens of examples of tech spreading faster.
this is the largest stockholder in AMAZON and a large stock holder in AI companies explaining why AI is industrial so its not like the housing bubble.
But Jeff, by that logic .com was industrial. It made everyone more productive etc etc. So, he cherry picked his comparison. Objectively, Bezos is a biased participant. If theres a crash, that man will buy everything up on the cheap. It's actually beneficial for him if there is a crash. Because he's correct about the AI potential, and if a crash happens he can buy some IP for cheap cheap cheap.