Farzad: Terafab Is a Big Deal + Research Fab at Giga Texas + 300+ Cybercabs in Austin + Gavin Baker on $1T in AI Commitments

Farzad called Elon's Terafab comments a big deal and said the result of the Research Fab at Giga Texas will hopefully be extremely high intelligence at extremely fast speeds at extremely low costs. He also RTd robotaxi on 300+ Cybercabs serving Austin, plus Gavin Baker, Silicon_Data and cljack on AI economics.

Terafab and the Research Fab at Giga Texas

SawyerMerritt, Oct 10 evening CT, posted a condensed version of Elon's Terafab remarks at X Takeover: "I still have a lot to learn about chip making. Gotta start somewhere. I didn't know how to make cars or rockets. The (chip) equipment's not very customized, it's all coming from the same vendors. If you tour a Tesla factory, we got tons of custom equipment." In the clip, Elon's full words were: "We're going to start with the conventional equipment, and then we're going to modify that equipment to improve the throughput. And then ... where we see the choke points, I think we'll make customized equipment that allows for a much higher throughput of wafers." Farzad liked it and quoted it, Oct 10 evening CT: Big deal.

TeslaHype, Oct 10 evening CT: Elon says the chip industry is risk-averse, so Tesla and SpaceX will make high-risk, high-payoff bets. The Research Fab at Giga Texas puts logic, memory, packaging and masks in one building for a super fast iteration loop. That's TeslaHype's summary of Elon. Farzad liked it.

Farzad quoted it, Oct 11 morning CT: The result of this will (hopefully) be extremely high intelligence at extremely fast speeds at extremely low costs. That's Farzad's projection, and the hedge is his.

Check Big deal QT here. SawyerMerritt Terafab here. TeslaHype here. Research Fab QT here.

300+ Cybercabs in Austin

Farzad RTd and liked robotaxi, Oct 10 afternoon CT. robotaxi said, Oct 10 afternoon CT, that one month after launch it has scaled to 300+ Cybercabs serving Austin, an 8x increase in 30 days.

Farzad liked aelluswamy, Oct 10 afternoon CT, quoting that post: Two years since the 10/10 Cybercab WeRobot event, the cabs are everywhere in Austin now. Dallas soon.

Farzad liked Elon, Oct 11 morning CT, also quoting it: Cybercab is scaling

Farzad liked mikepat711, Oct 10 afternoon CT: Having this ICE rental is making me so bullish on Tesla. Not having FSD is like having no AC. Europeans won't understand any of this post

Check robotaxi RT here. robotaxi here. aelluswamy here. Elon Cybercab here. mikepat711 here.

Gavin Baker on $1T in "commitments"

Farzad RTd and liked GavinSBaker, Oct 11 morning CT. Gavin, Oct 11 morning CT: No one - and I mean literally no one in the AI supply chain - takes the ridiculous $1 trillion of unfunded spending "commitments" from OpenAI and Anthropic even vaguely seriously. Hence the TSM executives dismissing Sam Altman as a nontechnical "podcast bro" after he came to them to make a commitment. Just listen to the answers that public company executives give about these unfunded "commitments." They are functionally irrelevant and this should be obvious to everyone. Those claims are Gavin's.

Gavin was replying to BetterIRR, a few minutes earlier. BetterIRR wrote about blowing up the 11th and 13th largest companies on earth with 1T in spending commitments and wondered whether the half trillion in non lab VC AI money goosing demand will continue at the same pace.

Check Gavin RT here. GavinSBaker commitments here. BetterIRR here.

Open-weight models, GPU prices, "the most important chart"

Farzad RTd and liked another GavinSBaker post, Oct 11 morning CT. Gavin, Oct 11 morning CT: Open-weight models that decrease margins at the model layer are positive for AI infra, exhibit 1001.2. Since the @Silicon_Data "token index," which captures both price and mix, peaked on May 31: B300 prices are up 38%, B200 prices are up 10%. Other GPU indices are up much more. The numbers are Gavin's, citing Silicon_Data's index.

Farzad RTd and liked Silicon_Data, Oct 11 morning CT. Silicon_Data, Oct 10 late afternoon CT: Allegedly "the most important chart" for understanding the current state of the AI economy. The model layer seems to be going through a period of commodification as non-frontier models saturate everyday use while frontier models compete fiercely. Compute layer sees massive increase in total demand for compute regardless of the source or use case. That read is Silicon_Data's.

Check Gavin open-weight RT here. GavinSBaker token index here. Silicon_Data RT here. Silicon_Data here.

Also

Farzad RTd and liked cljack, Oct 11 morning CT. cljack, Oct 10 evening CT: Talked to a finance guy today who said his firm has not only totally stopped buying software in favor of building it all in-house, they've actually hired multiple SWEs to do the buildouts, even though their salaries cost more than the subscription fees they were paying. Having tools custom-built for their specific business is just that valuable. So they're not spending less money or saving time, versus the pre-AI world, but they're getting a much higher return on the money and time they are spending. Much to ponder

Farzad RTd and liked tobi, Oct 10 late morning CT. tobi, Oct 10 morning CT: It's feeling really magical inside of Shopify right now. It's clearly the greatest time in the history of tech industry right now. I hope it feels similar everywhere else.

Check cljack RT here. cljack here. tobi RT here. tobi here.