Thing I realized: if GenAI really is the game-changer the fondlers say it is, it should have broken through to the consumer now, in the form of better goods and services at lower prices. But that’s not what I’m seeing. So either GenAI doesn’t work as advertised, or the companies deploying it are keeping the surplus profits to themselves. Neither scenario makes me want GenAI more.

I’m too fucking old to care about hipster terminals, so I had no idea ghostty was started by a (former) billionaire. If forced to choose a new terminal I will certainly take this fact into consideration.

the base use for LLMs is gonna be hypertargetted advertising, malware, political propaganda etc

well the base case for LLMs is that, right now

the privacy nerds won’t know what hit them

u wot m8???!!

Show HN: I analyzed 8k near-death experiences with AI and made them listenable

psychic damage warning, obviously

A lobster wonders why the news that a centi-millionaire amateur jet pilot has decided to offload the cost of developing his pet terminal software onto peons begging for contributions has almost 100 upvotes, and is absolutely savaged for being rude to their betters

https://lobste.rs/s/dxqyh4/ghostty_is_now_non_profit#c_b0yttk

This would be worrying if there was any risk at all that the stuff Anthropic is pumping out is an existential threat to humanity. There isn’t so this is just rats learning how the world works outside the blog bubble.

Apparently we are part of the rising trend of AI denialism

The rise of AI denialism

Author Louis Rosenberg is “an engineer, researcher, inventor, and entrepreneur” according to his PR-stinking Wikipage: https://en.wikipedia.org/wiki/Louis_B._Rosenberg. I am sure he is utterly impartial and fair with regards to AI.

This looks like it’s relevant to our interests

Hayek’s Bastards: Race, Gold, IQ, and the Capitalism of the Far Right by Quinn Slobodian

https://press.princeton.edu/books/hardcover/9781890951917/hayeks-bastards

HN discusses aliens https://news.ycombinator.com/item?id=46111119

“I am very interested.”

Bet you are, bud.

That popular piece on why it’s dumb to build (GenAI-scale) in space hit lobste.rs, and while most commenters agreed it is indeed dumb, fascist Flask founder felt the need to “well ackstually” for some stupid reason

https://lobste.rs/c/mmq7sy

duh ofc there are computers in space, there are computers everywhere, but the whole fucking point of the piece is you can’t take thousands of racks of GPUs and launch them into space and expect them to work

oh yeah the relationship between the fusion-device wielding 30-something Aluetian freedom fighter and the 16 year old skateboard courier in Snow Crash is… of its time

coming soon to whatever remains of CBS with an introductory 1 hour lecture by Bari Weiss

Neal Stephenson comes out strong and funny against GenAI here:

https://nealstephenson.substack.com/p/a-remarkable-assertion-from-a16z

“Hypothesis 1: it was written by a clanker”

I’ve worked in an adjacent field (workforce planning) and I deliver B2B software support for a living, so I too have Thoughts.

At least here in Schwedenland, contact centers have been filed down by relentless cost and tech pressure to be about as automated as can be. You have websites with FAQs, simple chatbots that basically repeat the FAQ for those for whom reading more than a sentence of text is too hard, phone trees to gatekeep you from the Inner Sanctum, etc. etc. The end result is that the actual people taking the calls are gonna be the ones who can make human decisions - troubleshoot a complex issue, handle insurance claims, upsell your mortgage.

Trying to att LLM voice tech to that is just going to add another filter between the customer and the center, with the additional reputational risk of the robot fucking up and losing the customer.

There’s base-level sneer as this is posted on LW, but I found this comparison of LLM in call centers to cheap human labor for the same interesting:

https://www.lesswrong.com/posts/rJatmEDcYrDQcwstT/the-economics-of-replacing-call-center-workers-with-ais#comments

He’s also found a lucrative niche being the “serious” AI booster. No doubt he gets something from it, if nothing else consulting work.

FFS so many promptfondlers and gish-gallopers in there. Echoes of pro-crapto (can’t criticize if you don’t buy in, use case is coming bro, it’s actually decentralized)

Edit the worst thing isnt’t the number of fondlers, it’s the upvotes they’re getting.

“Enjoy” this toxic stew of prediction markets, racism, and the objectifying of women:

https://protos.com/polymarket-criticized-for-racist-post-targeting-fake-baddies/

Saw this quality nickname for prompt receivers: Chet Jippity

Linux puritans are as fissile as the OG Puritans. There’s already Devuan which is Debian minus systemd, and now with XLibre. Not sure why this nerd thinks his particular non-denominational chapel will bring in the crowds to hear him sermonize.

Edit BTW Devuan for me is the epitome of FLOSS techfash.