I’m worried that “doom” speculation will make doom more likely. Specifically, AIs conform to our expectations of them, as communicated by their training data. This “self-fulfilling misalignment data” may be poisoning training already. 🧵

The 5 levels of #Linux users: Level 1 - Ubuntu Level 2 - Manjaro Level 3 - Debian Level 4 - Arch Level 5 - Ubuntu I reached Level 5. And you?

In 20 years, NIH leadership and universities have done literally nothing to address this issue. If you just try to recreate experiments. You can only do it 10 -40% of the time. No one stood up for reproducible science

What the actual fuck? USA's real moat is global talent -- and Chinese engineers are among the most craked, hard-working people at FAANG and top AI labs. Is the US seriously prepared for more DeepSeek, UniTree, DJI, and AI chips emerging from China?

Elon creates enormous value for historians with a concrete, tweet-by-tweet demonstration of how a tremendously successful man loses contact with reality and locks himself in an echo chamber of court eunuchs. He seems to believe everyone except some leftoids still likes him

piss-everyone-off take: China should be a democracy. They are a homogenous, docile, sane population. They'll outperform the CCP. ≈30% of the US is wannabe gangstas, shart-in-the-mart biofuel & Tesla burners. Democracy means their “opinions” drag the country down. Xi is better.

Id be crying if I was like that too, bruh, that's fucked up what NERV be doin' to you.I ain't even gon hold you, I be sayin' that's fucked up. Like bruh, you prolly had the full life, your shit be fire prolly if they ain't make you pilot EVA. But fuck it tho, bro, it's your life.

Geoffrey Hinton says RLHF is a pile of crap. He likens it to a paint job for a rusty car you want to sell.

I don’t have anything novel to contribute on the substance of ai-2027.com but have to again comment, pace Situational Awareness that I think kicked this trend off, that single-essay microdomains with a bit of design, a bit of JS, and perhaps a downloadable PDF are…

A new legal letter aimed at OpenAI lays out in stark terms the money and power grab OpenAI is trying to trick its board members into accepting — what one analyst calls "the theft of the millennium." The simple facts of the case are both devastating and darkly hilarious. I'll

AGI is the most important and potentially dangerous technology of our time. OpenAI was right that this technology merits strong structures and incentives to ensure it is developed safely, and is wrong now in attempting to change these structures and incentives. We're urging the

Today, we Iranians are faced with something truly unbelievable: Western, liberal, “woke” feminists, clueless about the brutal reality in our country, consistently stand with our oppressors under the delusion of saviorism. They have no understanding of the decades of humiliation,

I think that people who assert that current AI models definitely don't have subjective experience are trying to signal that they're 'serious people'. But to me it really just reads as an embarrassing tell that they're completely unaware of the relevant philosophy.

Opposition to lab grown meat is just mindless hysteria. If we can recreate the taste and texture of meat without any of the associated negatives, people should be happy. If we no longer have to raise and slaughter billions of animals a year that would be great for our planet.

it’s happening

Nick Bostrom’s new paper: >Developing superintelligence is not like playing Russian roulette; it is more like undergoing risky surgery for a condition that will otherwise prove fatal. > One could equally maintain that if nobody builds it, everyone dies. In fact, most people are

It happened As expected Scam Altman sold out again all in the name of the mission "to serve all of humanity" I expect 89 resignations by end of day

"You need to understand that Sam can never be trusted." "He is a sociopath. He would do anything." - Aaron Swartz on Altman, shortly before he took his own life