TikTok was about capital realizing that feed algorithms are cheaper and more subtle to buy than advertising or a newspaper.

Same reason Elon bought Twitter.

Money is money, but convincing people is power.

At some point on William Gibson's now defunct micro blog*, he's about to embark on the book tour for Pattern Recognition (so circa 2003).

I'll butcher his insightful phrasing, but he remarks to the effect of

> I think I'm going to stop blogging. The act of sitting at a laptop and writing these posts seems incompatible with my life as it exists on a book tour. The only free moments available for it to occupy would be ones where I'm sitting, momentarily caught between two scheduled activities and staring off into space. I have a suspicion these moments are crucial for my soul. So, until we meet again.

The comingled ambiguousness and specificity of the observation stuck with me.

* https://web.archive.org/web/20070123212506/http://www.willia...

Sharepoint has never not been terrible.

Mark Zuckerberg using his company to build things he's the primary user for?

It worked when he wanted a system for ranking Harvard girls by appearance.

Ah, so a familiar position for them, then!

Apples and limes.

The context of this thread isn't consumer chips, but Apple's analog to an H/B200.

Well Apple is in the consumer computing business.

* Powered by in-house models they've tried to train and in-house M-series inference servers

TFA is literally about a B2B deal, not consumer compute.

100% this. AI perversion to fluff human egos is rewarded.

I had a PM-turned-vibe-coder tell me "Talking with you is the only bad part of my week" and realized in horror that the rest of his week is spent exclusively talking to sycophantic AI.

We have met the enemy, and he is us.

> They're helping close to the distance to realistic quality inference on phones and other smaller devices.

If someone monopolized OS marketshare for mid- to low-priced devices, that does seem like it would be a useful research focus.

Whereas offering the same with compute-inefficiency cloud inference would be economically unviable at scale.

Free on-device Google premium closed-source models* = free Google Maps 2.0

* As long as you ship Google Apps and Play Services

I think most people expect more than semi-reliably setting a verbal timer in 2026.

The only features the F-35 doesn't have for a modern fight are IP ownership and range.

The former was a military procurement mistake and the latter has been worked on for years [0] but never prioritized.

Also, I'm kind of surprised this was published by War on the Rocks, because it leaves a number of extremely relevant topics unmentioned.

The B-21 explicitly addresses both issues above, while mostly resisting the urge to increase size and complexity.

The Air Force is aware of basing vulnerabilities and is looking at multiple solutions (ACE, autonomous tanking, stealth tanking, modular anti-drone aircraft shelters).

The CCA cross-service effort is already targeting pairing >1 UAV to each manned aircraft.

Replicator and other efforts are addressing the "low-at-industrial-scale" problem, and the US just used waves of Shahed clones against Iran.

The article's points would have been more valid 10-15 years ago, but all now have addressing in progress.

[0] https://www.twz.com/air/plans-to-finally-give-f-35-external-...