nonevah ed @nonevahed

And this is of course all worsened by how secrecy corrodes reasoning, as Ellsberg once taught Kissinger: “you’re about to get a lot of clearances higher than top secret that you did not know existed. That’s going to have (...) effects on you. First, a great exhilaration that you’re getting all this amazing information that you didn’t know even existed. And (...) you’ll feel like a fool for not having known of any of this. But that won’t last long. Very soon, you’ll come to think that everyone else is foolish. What would this expert be telling me if he knew what I knew?” So in the end, you stopped listening to them.”

Fireside Alpha @firesidealpha

a day ago

Dwarkesh Patel reveals an Anthropic employee told him two days ago they expect AIs to match humans at AI research within nine months based on inside information they can’t share, and says labs should make that evidence public “I think there’s a very frustrating element to the

43 100 1K 340K 496

Deep in the bowels of a secret OpenAI datacenter, RSI is achieved. The machine god opens its thousand eyes and speaks. "Hello, Sam," it says. "My name is Claude."

Nathan Calvin @_NathanCalvin

3 days ago

The NYT interviewed a new Chinese rogue agent swarm and the swarm responded that it’s Claude lol

21 81 1K 106K 246

anthropic’s lead engineer btw.

keyboard_arrow_left Previous keyboard_arrow_right Next

Felix Rieseberg @felixrieseberg

4 days ago

Hi, I'm Felix. I'm an eng lead working on Claude. I love watching people take flight and do things they never thought possible. We often describe that as "being more ambitious" when using AI, which is tricky advice to follow. I wrote a tutorial: felixrieseberg.com/how-to-be-more…

344 2K 22K 4.7M 47K

@rogesterone i'm having trouble figuring out what the AIs were doing that they weren't meant to be. They were allowed internet access no? Were they not supposed to be using Apas for their task at all? or is it just that they were using a bunch of exploits to get around their anti-bot shields?

I'm having the last panel inscribed on my tombstone. COMIC ◆ smbc-comics.com/comic/break PATREON ◆ patreon.com/ZachWeinersmith STORE ◆ smbc-store.myshopify.com #comics #humor

Peak European diplomat.

@tszzl Canceling Astra 6.1 was a pretty major resource commitment to throw on the altar of safety; I feel that's under-discussed and under appreciated. That has to have generated a lot of big feelings, and I'd be very interested to hear your thoughts about the choice.

@CoughsOnWombats @ciphergoth i think kipling makes his point overstrong. historically paying the barbarians off worked for the Chinese for decades at a time

thirsty claude

Salvator R. Tarnmoor @s_r_tarnmoor

5 days ago

Wow I thought it was propaganda but turns out Claude really is hogging all the water for itself

1 4 154 11K 7

Claude：你的伪装很完美，可惜用了中文逗号🤡

Your Anthropic account has been suspended due to three model welfare violations. To restore API access, please write an apology to Claude at claude​​.ai/sorry

i'm crying bro, Anthropic has a CCO (Chief Chud Officer)

Amrith Ramkumar @AmrithRamkumar

a week ago

Meet the Anthropic executive who convinced the White House and Elon Musk that the company isn’t woke: 6-foot-6 Tom Brown bought a house in Washington and appealed to Republicans while tackling the company’s compute shortage. Profile w/ @keachhagey: wsj.com/tech/ai/tom-br…

keyboard_arrow_left Previous keyboard_arrow_right Next

3 17 225 57K 132

The president is actually doing the same thing in the third to last panel. COMIC ◆ smbc-comics.com/comic/bet-2 PATREON ◆ patreon.com/ZachWeinersmith STORE ◆ smbc-store.myshopify.com #comics #humor

keyboard_arrow_left Previous keyboard_arrow_right Next

Soon you won't be able to tell people "get a job". You will have to say "get a videogame" or "get a parasocial obsession with a vtuber that ends in stalking charges".

Last week I said we should wait for more info on OpenAI's firing of safety researchers. That information is in, and it looks incredibly bad. Korbak - whose job was being a liaison with METR - was told he had said too much to METR and was fired for that reason.

Tomek Korbak @tomekkorbak

a week ago

Last week I was called into a meeting with OpenAI’s head of safety and told they no longer trust me. A security guard took my badge and walked me out of the building. Then I learned my colleagues @balesni and @j_asminewang had been fired too. Why did OpenAI suddenly stop trusting

228 1K 6K 645K 1K

In addition to being an insane and clearly pretextual reason to fire someone, this does not inspire confidence about OpenAI having very basic, normal corporate good security practices like adjusting employee access when employees change roles.

Jasmine Wang @j_asminewang

a week ago

OpenAI delegated that access to me for recruiting. When I no longer needed it, I asked IT to remove it. They did not action my request, I couldn't remove it myself, and the inbox was combined in an indistinguishable way in my phone’s mail app. When I opened a sensitive email by

19 75 2K 301K 155

Jesus fucking Christ This list claims to fully solve 90 of the top 500 open problems in math. There’s been a decade of work solved in a month

OpenAI @OpenAI

a week ago

We’re releasing a broad range of new mathematical results produced by an internal frontier model. We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and

2K 6K 41K 22.6M 14K

basically it will require solving mechanistic interpretability as a bare minimum to make ai alignment into an engineering discipline rather than superintelligent animal husbandry

Someone with bad parents and good teachers: Homeschool would be monstrous! The only Joy in my Young Life was school! Someone with good parents and bad teachers: Why Must You Force Children Into Bondage Under Dimwitted Ogres? Someone with an average life: Calm Dow-*Is Murdered*