Context: Post #4 in my sequence of private Lightcone Infrastructure memos edited for public consumption.

(emphasis mine)

Look at this goober, trying to be THE cult leader.

This. Masayoshi son is selling furniture to YOLO more money into OpenAI.

I have no opinion, but I have to note that I keep reading “KeepAssXC …”

Pivot to bio-computing

Beff Jezos and friends have produced something other than tweets. Possibly. Maybe.

https://youtu.be/Y28JQzS6TlE

Looks like a hobby project of someone who has very particular views about computers.

I’m not sure what kind of neural network are they planning to run on a custom FPGA based GPU with 4GB RAM shared with CPU

“How dare you suggest that we pivoted to SlopTok and smut because of money if something that we totally cannot do right now is more lucrative?”

If you use your business to “do business” it’s nice to have good catering.

Simon Willison writes a fawning blog post about the new “Claude skills” (which are basically files with additional instructions for specific tasks for the bot to use)

How does he decide to demonstrate these awesome new capabilities?

By making a completely trash, seizure inducing GIF…

https://simonwillison.net/2025/Oct/16/claude-skills/

He even admits it’s garbage. How do you even get to the point that you think that’s something you want to advertise? Even the big slop monger companies manage to cherry pick their demos.

Just felt like I got an aneurysm there.

(in unrelated things, first)

They will all be just this one:

https://c.tenor.com/-MgJ9BsDFvwAAAAC/tenor.gif

Slack CEO responded there that it was all a “billing mistake” and that they’ll do better in the future and people are having none of it.

A rare orange site W, surprisingly heartwarming.

I call it: “scientific” racism. To me that gets the point across the best.

I wondered if this should be called a shitpost or an effortpost, then I wondered what would something that is both be called and I came up with “constipationpost”.

So, great constipationpost?

Was jumpscared on my YouTube recommendations page by a video from AI safety peddler Rob Miles and decided to take a look.

It talked about how it’s almost impossible to detect whether a model was deliberately trained to output some “bad” output (like vulnerable code) for some specific set of inputs.

Pretty mild as cult stuff goes, mostly anthropomorphizing and referring to such LLM as a “sleeper agent”. But maybe some of y’all will find it interesting.

link

It’s two guys in London and one guy in San Francisco. In London there’s presumably no OpenAI office, in SF, you can’t be at two places at once and Anthropic has more true believers/does more critihype.

Unrelated, few minutes before writing this a bona-fide cultist replied to the programming dev post. Cultist with the handle “BussyGyatt @feddit.org”. Truly the dumbest timeline.

Yeah, didn’t even cross their mind that it could be wrong, because it looked ok.

Shamelessly posting link to my skeet thread (skeet trail?) on my experience with an (mandatory) AI chatbot workshop. Nothing that will surprise regulars here too much, but if you want to share the pain…

https://bsky.app/profile/jfranek.bsky.social/post/3lxtdvr4xyc2q

That’s how you get a codebase that kinda sorta works in a way but is more evolved than designed, full of security holes, slow as heck, and disorganized to the point where it’s impossible to fix bugs, adds features, or understand what’s going on.

Well, one of the ways *glancing at the code I’m responsible for, sweating profusely*

resulting in one person getting bacon added to their ice cream in error

At first, I couldn’t believe that the staff didn’t catch that. But thinking about it, no, I totally can.

He’re hoping that the more he spends time gooning the more he’ll leave the rest of us alone. crosses fingers