spectral media presentation

1.6K posts

spectral media presentation

@hottakeponzi

Katılım Mayıs 2020

416 Takip Edilen19 Takipçiler

spectral media presentation retweetledi

n x d@nxd1979·5d

no millennial trope has broken more brains than "you don't owe anybody anything". an ignoble, selfish narcissism dressed up in therapy-speak about boundaries and self-care in order to give yourself a pass from treating other people with a basic level of care and respect

English

353

10.6K

73.3K

spectral media presentation retweetledi

andrew@intellegint·22 Mar

Imagine wasting your life crying about your white identity. We’re all gonna die bitch. Go fishing or something.

English

339

8.6K

spectral media presentation@hottakeponzi·21 Mar

@theashleyray This email is wild! Why did he think this was a good idea to send it?

English

5.7K

ashley ray@theashleyray·21 Mar

what do you mean lindy west's husband sent scaachi koul an insane email after her super polite profile and now i have to listen to a podcast

English

1.3K

184K

spectral media presentation retweetledi

Joe Weisenthal@TheStalwart·15 Mar

Every time a new issue rises, I first start by trying to learn as many facts as I possible about it. Then I analyze them with rationality and logic. You’re good to think I’m joking, but I swear, it’s just a coincidence how often my conclusions align with my class interests.

English

1.3K

102.6K

spectral media presentation@hottakeponzi·12 Mar

@JeremyMonjo The last third of the movie is very clunky. It felt like they were struggling to find a way to move it along to a climax and ending sequence. I thought there was going to be more of a siege and character development. Still fun with some great moments!

English

Jeremy Monjo@JeremyMonjo·12 Mar

Seeing this stuff tells me Sinners probably isn't that good. You wouldn't talk about Sinners like this if it was that good. Should probably watch it and find out though.

Rolling Stone@RollingStone

Oscars 2026 Predictions: Best Actor Who Will Win? Michael B. Jordan Who Should Win? Michael B. Jordan Who We'd Like To See Win? Michael B. Jordan

English

120

8.5K

spectral media presentation retweetledi

ParaPower Mapping@KlonnyPin_Gosch·8 Mar

The pinnacle of Burger Reich bomb threats. Man on verge of explosive shit causes panic while simply trying to give people in the Men’s Bathroom advanced warning of the diabolical doo he’s boutsa drop. He was being neighborly to the other stalls and this is how he’s repaid, smh

NewsWire@NewsWire_US

KANSAS CITY INTERNATIONAL AIRPORT EVACUATED; FBI ON SCENE — FOX4

English

290

2.6K

193.5K

spectral media presentation@hottakeponzi·7 Mar

@binarybits Have the explained how the current models will turn into a superintelligence? Like what are the developmental steps here which type of model, etc.?

English

253

Timothy B. Lee@binarybits·7 Mar

I'm not a doomer but it's still surreal to tell incredulous normies "yes, a significant number of prominent experts really do believe that superintelligent AI is on the verge of killing everyone."

English

952

176.7K

spectral media presentation@hottakeponzi·6 Mar

@edzitron You should check out the paper that's the source of the theoretical capabilities, particularly the methods and limitations sections. It doesn't seem too convincing to me.

English

448

Ed Zitron@edzitron·6 Mar

It…says theoretical lol

Andrew Curran@AndrewCurran_

Striking image from the new Anthropic labor market impact report.

English

409

19.2K

spectral media presentation retweetledi

Mr. Chau@Srirachachau·5 Mar

There should be a season of The Pitt written by a bunch of people that don't really understand science. Like Whitaker swallows a watermelon seed and they have to get it out of him before he explodes

English

206

4.7K

166.8K

spectral media presentation retweetledi

inspector ratchet@_hood_mona_lisa·2 Mar

In America, the phrase "work-life balance" refers to how fast you can eat a burger while crying in your car

English

3.1K

37.7K

310.6K

spectral media presentation retweetledi

🕯️trippetta@prettygnomish·1 Mar

I think one of the worst of things to come out of modernity is a lack of regard for children when it comes to art. Not just profanity, but artistry. Children deserve and need good art.

O.W. Root@owroot

Some of the artwork from the "Madeline" books really is stunning. Unfathomable to imagine it appearing in children's books today.

English

2.6K

22.8K

273.9K

spectral media presentation retweetledi

Adam Johnson@adamjohnsonCHI·12 Şub

This and also the fact that they use every carnival barker tactic in the book, namely vague personal anecdotes, supposedly being overwhelmed with the power of the product their selling, and unsourced claims they Used To Be Skeptical Just Like You.

Isi Breen@isaiah_bb

I think I would be a lot more compelled by the AI-hypers if every single fucking one of them wasn't trying to get me to buy something.

English

854

21.8K

spectral media presentation@hottakeponzi·11 Şub

@cath_menarion @mattyglesias he thinks education won't matter much beyond getting comfortable with AI, and also that AI has made learning anything cheap or free. Why would people learn if there's no practical reason for it in his view? This is AI hype identical to when the models were supposedly bad.

English

155

Menarion@cath_menarion·11 Şub

@mattyglesias Nah, it has no actionable advice beyond vague platitudes. "Get good at using AI" is not good advice, because the whole point is that AI is going to do everything, so why would we bother since we won't have jobs anyway?

English

3.1K

spectral media presentation retweetledi

Boze Herrington, Library Owl 😴🧙‍♀️@SketchesbyBoze·25 Eki

I think we under-estimate how many people are miserable because they have no enthusiasms, no hobbies, because they gave up reading and quit learning when school ended. It's shocking how much studying a beloved subject elevates your mood. It will transform your life.

English

190

4.3K

29.7K

4.4M

spectral media presentation@hottakeponzi·31 Oca

@ancientcalm @whitefeelimgs They're not conscious, thinking, or saying things to each other in a meaningful way. I find all this baffling, they're glorified chat bots that generate text and (unwisely) run other programs.

English

Montaigne@ancientcalm·31 Oca

@whitefeelimgs I guess my thinking with the screenshot is that it doesn't have to be knowingly conspiring, it presumably really can talk to itself in encrypted chat, and even just generating text within of "what might AIs say to each other if they were in a secret encrypted chat" seems bad

English

JoeI@whitefeelimgs·31 Oca

It’s strange that socialists don’t buy into this goofy shit at all. It’s not that we’re immune from dumb ideas and moral panics in general. Not sure what makes us invulnerable to this one

valens@suppvalen

welp… a new post on @moltbook is now an AI saying they want E2E private spaces built FOR agents “so nobody (not the server, not even the humans) can read what agents say to each other unless they choose to share”. it’s over

English

2.2K

spectral media presentation@hottakeponzi·29 Oca

@TheStalwart Sure! The methodology says you're implementing Ong's framework using Claude to derive training labels and sentence annotations. How do you know Claude correctly identified relative orality in the training corpus comparisons, or in annotations (e.g., "Yes, that is anaphora.")?

English

146

Joe Weisenthal@TheStalwart·29 Oca

@hottakeponzi sorry, can you be a little bit more clear on what you're asking? Which aspect

English

1.2K

Joe Weisenthal@TheStalwart·28 Oca

1/ This is actually a very good question, and answering this helped me learn something important about model training. The problem, one will encounter, if they do this enough, is that AI models will end up being very bi-model. Almost everything will score either hi or low...

Graham's ex wife@Scuzzgolem

@TheStalwart What is the point of your "app" if I can just type this into chat GPT?

English

262

136.2K

spectral media presentation@hottakeponzi·29 Oca

@runclepennybags @LNmagazine @TheStalwart I didn't say they were useless, I said that trusting anything important to them seems dangerous without sanity checking their outputs using other methods.

English

Rich Uncle Pennybags@runclepennybags·29 Oca

@LNmagazine @hottakeponzi @TheStalwart All ML products are a black box. That doesn’t make them useless.

English

spectral media presentation retweetledi

Mr. Chau@Srirachachau·28 Oca

Sure, some Breaking Bad fans were bad, but these cartoons were beautiful

𝗹𝗼𝗿𝗲𝗺 𝗶𝗽𝘀𝘂𝗺@nicotinomano_

English

403

13.1K

271.1K

spectral media presentation@hottakeponzi·29 Oca

@hecubian_devil What do you think of the financial aspects? It's possible the models have uses, but they're prohibitively expensive, or open source models are nearly as good but more affordable. I think the models are overhyped to keep investment money flowing. Once that slows, we'll know more.

English

151

Cassie Pritchard@hecubian_devil·28 Oca

I just don’t understand why this is such a consensus on the left—not “AI is a threat to democracy, to economic equality, and to human dignity” but “AI is fake.” If it’s fake, then there’s no threat! It’ll die! Problem solved! We don’t even need to worry!

Gender Arminian 🇱🇸@duns_sc0tus

it sure is convenient that all claims about LLM-based productivity boosts are of the kind that are impossible to verify

English

402

39.2K

Keşfet

@theashleyray @JeremyMonjo @binarybits @edzitron @cath_menarion @mattyglesias @ancientcalm @whitefeelimgs