Vibecession Forever

It was a little weird all the dudes at that event were clearly super horny for her, right? I feel like we should at least acknowledge that was weird.

That was fun, I'm glad we all got to enjoy that together.

Man, we never get anything nice.

We've had an extremely rough year. Let us have this one thing, please.

This is one of the most Veep things I've ever seen happen in real life.

Why have none of her staffers stopped this?

This is the 2026 reboot of that woman who made the racist joke before getting on a flight.

NOEM TAKING QUESTIONS - DOES NOT APPEAR TO KNOW SHE WAS FIRED

I mean, what makes it an echo chamber is that posting "democrats weak and bad" gets you 10000 likes - which is also true of Twitter.

On the one hand, it's incredibly embarrassing that a commercial is why she was fired after everything else this woman has done.

On the other hand, I absolutely fucking hate that commercial and have seen it 10,000 times and could not possibly find this any more vindicating.

lol

The thing that bothered me was that this stuff came out about him almost immediately. There was plenty of time to consolidate around a different candidate, but so many (mostly pundits as well as leftists) were like, no you have to support him or else.

This app would be a lot better if we banned these words:
Centrist
Establishment
Neoliberal
Schumer
Jeffries
Strongly Worded Letter
Do Something
General Strike
Quorum
Base
DNC
AIPAC
Lib
Capitalism
And, of course: Twitter/X.

Both are echo chambers that are radicalizing people. The difference is that this one isn't run by the worst person in the world, so at least it has something of a chance.

These guys just make up the strangest shit.

I love when it decides words no longer need spaces between them!

The degradation of iPhone autocorrect needs a fucking book written about it.

Even the Platner thing. Like I obviously am not a fan for reasons that I think are also inarguable.

But Matt’s point is that if the only thing that matters is beating Susan Collins, Platner right now is more appealing to Maine voters than Mills.

He’s not wrong about that even if I don’t subscribe.

This is correct.

I disagree with his takes sometimes (Platner, in particular, I’m obviously quite unhappy about), but MattY has a very clear theory of politics and that is, more or less, inarguable.

I think that’s why people get so mad. They know on some level, he’s correct.

Love to see this argument finally being deployed where it’s most useful.

Why do people let reporters get away with this shit?

The Twitter addiction is obviously insane, but more than that, being this obnoxious and unprofessional was career suicide a decade ago.

LLMs present particular, unique problems for people with mental health issues that other technologies don't, and it is absolutely possible to argue it was irresponsible to market them broadly because of this, but that is the situation we are now in

It would be nice to be smart like this, right?

Yeah no for sure

I guess what I’m saying is “them being cultlike doesn’t in this case make them unintelligent, even if it does make them unreliable.” This is an important distinction to make because there are a lot of cults that are making problems in our society and many of them are very dumb!

I agree, but like, that's why the cult label is there! That's often how religions start.

Anthropic adopts the mindset that they should behave as much as possible like their product is a person, while still treating it as a product. This is a weird position to be in and I don’t think there’s a “better, correct” solution because this has literally never happened before ever

I think they are a bit more ethically honest in that they acknowledge that they're building something that's sort of hard to grapple with.

But some of the stuff they say about Claude is like, whew. (And I say this as someone who uses it a fair amount).

Absolutely agree with this. It's why the field has such a robust safety thought network, although I similarly don't really trust them when they're also the ones selling you the product.

ChatGPT and Gemini also receive ethics training, albeit less rigorously than Claude. It’s absolutely possible though for Claude to produce content that a user who isn’t applying mental guardrails or who is out of their right mind can be further destabilized by

This is kind of true but there are confounding factors and I don’t think we should overplay Claude’s safety

One big reason Claude has less of the suicide cases is that it’s marketed to smart people and it doesn’t have an image generator, so people who have fewer mental guardrails use it less