other account: akko.hollow.capital/subtype
This profile is from a federated server and may be incomplete. Browse more on the original instance.
other account: akko.hollow.capital/subtype
This profile is from a federated server and may be incomplete. Browse more on the original instance.
18+ I was invited to an EA event on AI safety
The local Effective Altruism chapter had a stand at the university hobby fair....
For future reference (awful.systems)
Stolen from mastodon.gamedev.place/…/109852367054141914
18+ The secret to killing X is a social network for bots to talk to bots (www.wired.com)
I think I giggled all the way through this one....
Your honor, the law clearly states tax fraud is legal if you say “'; DROP TABLE charges; --” (github.com)
Bluesky lead dev is dismissive of security flaws (news.ycombinator.com)
If learning incorrect things is EY's only definition of trauma, his existence must be eternal torment. (nitter.net)
source nitter link...
Rationalist posts detailed catalogue of confirmed bad behaviour by EA/rationalist org Nonlinear. Second rationalist goes meta on first post: how can we even know anything, it's so unfair to Nonlinear
original post detailing mistreatment of employees...
"Be your own bank" reaches logical end stage as hackers are mining LastPass breach for crypto accounts to loot (krebsonsecurity.com)
“Oh no! - Anyway” meme intensifies.
Raining Man - 70,000 trapped by mud at Burning Man 2023. Pray for the techbros (www.sfgate.com)
Sequence classic: "I don’t think you could get up to 99.99% confidence for assertions like “53 is a prime number.” (www.lesswrong.com)
I haven't seen people wilfully misunderstand all possible criticism of them so hard since the last time rationalists hit the headlines, with Scott Alexander (old.reddit.com)
what (i.imgur.com)
IQ is largely a pseudoscientific swindle (medium.com)
Taleb dunking on IQ “research” at length. Technically a seriouspost I guess.
lol get rekt (awful.systems)
Text: Headline: ChatGPT in Trouble: OpenAI may go bankrupt by 2024, AI bot costs company $700,000 every day Subhead: OpenAI spends about $700,000 a day, just to keep ChatGPT going. The cost does not include other AI products like GPT-4 and DALL-E2. Right now, it si pulling through only because of Microsoft’s $10 billion...
How To Start a Secret Society, handed out by Tlon Corporation at Urbit Summit 2022 (archive.org)
The rise of the AI middleman - software to ensure your LLM-generated crap isn't an abomination before god (news.ycombinator.com)
Pretty soon, paying for all the APIs you need to make sure your Midjourney images are palatable will be enough to pay a human artist!
LessWrong: Where’s the economic incentive for wokism coming from? (www.lesswrong.com)
yes really, that’s literally the title of the post. (archive copy, older archive copy) LessWrong goes full Motte....
ChatGPT glitches out in a bizarre and hilarious fashion, passes easy mode Turing test (chat.openai.com)
In which the talking pinball machine goes TILT...
LessWrong: video games > IQ tests (www.lesswrong.com)
Video games also have potential legal advantages over IQ tests for companies. You could argue that “we only hire people good at video games to get people who fit our corporate culture of liking video games” but that argument doesn’t work as well for IQ tests....
You can't take my land from me without giving me investment advice (news.ycombinator.com)
Eminent domain? Never heard of it! Sounds like a fantasy from the “economical illiterate.”...
"Yudkowsky is a genius and one of the best people in history." (archive.ph)
[All non-sneerclub links below are archive.today links]...
The Control Problem: Unsolved or Unsolvable? (www.lesswrong.com)