Friday: Three Morning Takes

A new bill that would legalize robotaxis in DC has faced resistance from a shadowy group called the People First Transportation Alliance. As WaPo reported, the PFTA lists several folks as “founding members” who told the paper they’d never heard of the organization… soooo: who exactly is behind this shadowy cabal of decels? Since Waymos cause about 63% fewer crashes per mile than human drivers, I suspect that the PFTA is funded by Satan or China — but per WaPo’s reporting, there are at least two loose ties between the org and Uber’s lobbyists. Well well well. I guess Uber, which has been losing some business to Waymo, is getting a little taste of what it’s like to be left behind. Dear Uber: instead of (potentially) funding the pro-car accident people, try working harder? To quote Travis Kalanick, the CEO you fired for being a bit much (lol), “Fear is the disease. Hustle is the antidote.”

In the latest hysteria over the campus rape case that just keeps giving, Columbia students are engaged in productive debate sending death threats to authors at the Columbia Sundial (proud founder here!) who argued it’s “worth taking a pause” before jumping to conclusions, and that it “infantilizes women” to assume Jane Doe had no agency. The “Columbia Open Discourse Alliance” cancelled a panel on *checks notes* press freedom, featuring Sundial’s editor in chief, in the wake of the op-ed. Said the president: “We did so not in spite of, but because of our commitment to free expression.” Also, war is peace. The meltdown at Columbia is a reminder that, on our nation’s elite college campuses, the Overton window on whether women can ever be responsible for the consequences of sexual liberation hasn’t budged. Remember, kids: the point of a $90k/year liberal arts education is open-mindedness, strictly within the designated 3-degree radius of acceptable thought.

Wednesday, Pirate Wires reported that SF, once a wasteland for uggos, is now teeming with beautiful women. On X, these women post viral selfies announcing their new jobs in AI sales and growth, while men we interviewed are clocking 10s in the Marina. Is this a sign we’re in… a bubble? AI startups, swimming in VC cash, are now shifting their energies toward buying attention, i.e. sexy saleswomen to help them stand out. Whether those eyeballs convert to customers is questionable, but as the debate continues over whether baddies in the Bay means the economy is about to tank, consider these examples of peak froth: one AI company hired a “Head of Coffee” (whaaaat?), another offers Porches to employees who refer new engineers, and a third pays staff $1,000 monthly to live nearby. Enjoy the blondes, booze, and cars while they last, friends (and maybe consider shorting the market).

Jonas Du

One Man’s Quest to Build a Very Small Data Center on a Remote Plot of Sand

strata founder david meyer recently embarked on a “listening tour” through rural colorado in hopes of building what might be the most reasonable data center project in america: a dozen shipping-container-sized boxes on a remote plot of sand. no water. no noise. the local electric co-op even said the project would lower power bills. you might expect the approval process to be fairly straightforward here... what followed was anything but. jonas shadowed david throughout the founder’s arduous quest, delivering this eye-opening, on-the-ground dispatch

Nicole Wizman

San Francisco’s Hot Girl Bubble

multiple sources reported to us that gorgeous women are flocking to the city of san francisco. nice! but before you celebrate a city of 5’s now being flooded with 9’s and 10’s, stop and consider: could this be a worrying sign for the economy? as nicole notes, perhaps companies that are flush with venture cash paying baddies to woo business prospects is... a pretty good sign we’re in a bubble. an important analysis (disclaimer: not financial/relationship advice)

G. B. Rango

Did This AI Research Breakthrough Just Avert the Apocalypse?

could we prevent AI from wiping out humanity if we... simply understood how it works? in a landmark essay, former openai researcher nick cammarata argued that understanding AI’s ‘reasoning’ is more possible than we previously thought. but does this mean it’s easier to control? safetyists like eliezer yudkowsky still aren’t convinced. our boy g.b. breaks down the new findings and the ensuing debate over AI “interpretability”