You are heroes (@1hopecomingtrue) 's Twitter Profile
You are heroes

@1hopecomingtrue

This account's AI Safety views are a mix weighted as follows:
55% Brockist 💾 + 30% decel ⏸️ + 15% e/acc ⏩
A recent Reddit thread explains twtr.to/tIv2q

ID: 1675025492393734146

calendar_today01-07-2023 06:16:27

867 Tweet

3,3K Followers

254 Following

Nicole Miller 🏛️ ‍⚖️ (@josourcing) 's Twitter Profile Photo

The Information Amir Efrati Jessica Lessin aaron holmes According to his sister, Sam Altman abused her in every way. To date, Satya Nadella and Microsoft have NOT severed their relationship with him nor removed his products from your computer. #anniematters lesswrong.com/posts/QDczBduZ…

Daniel Eth (yes, Eth is my actual last name) (@daniel_271828) 's Twitter Profile Photo

Good news from a safety perspective. I like Microsoft and think they’re more responsible than most companies, but this would have been a bad conflict of interest between Microsoft’s profit-seeking and OAI’s (justified) non-profit mission

AI Notkilleveryoneism Memes ⏸️ (@aisafetymemes) 's Twitter Profile Photo

What happens when you mix 1) open source AI (w/ no safety controls) 2) ~no regulation 3) autonomous agents proliferating uncontrollably? Internet lockdowns. The days of the open internet may soon be over as swarms of self-improving AIs wreak havoc, and quarantines are the ONLY

What happens when you mix

1) open source AI (w/ no safety controls)
2) ~no regulation
3) autonomous agents proliferating uncontrollably?

Internet lockdowns.

The days of the open internet may soon be over as swarms of self-improving AIs wreak havoc, and quarantines are the ONLY
U.S. Securities and Exchange Commission (@secgov) 's Twitter Profile Photo

The Commission has adopted a new rule intended to prevent the sale of asset-backed securities (ABS) that are tainted by material conflicts of interest.

Andrew Critch (🤖🩺🚀) (@andrewcritchphd) 's Twitter Profile Photo

Yann LeCun is calling the list of scientists and founders below "idiots" for saying extinction risk from AI should be a global priority. Using insults to make a point is a bad sign for the point… plus Hinton, Bengio, and Sutskever are the most cited AI researchers in history:

Yann LeCun is calling the list of scientists and founders below "idiots" for saying extinction risk from AI should be a global priority. Using insults to make a point is a bad sign for the point… plus Hinton, Bengio, and Sutskever are the most cited AI researchers in history:
James Campbell (@jam3scampbell) 's Twitter Profile Photo

🎉New paper up--introducing "Localizing Lying in Llama": We prompt Llama-2-70b to lie about T/F statements and localize where in the network this behavior occurs This project was led entirely by three undergrads: myself, Phillip Guo, & Richard Ren 🧵1/N

🎉New paper up--introducing "Localizing Lying in Llama":

We prompt Llama-2-70b to lie about T/F statements and localize where in the network this behavior occurs

This project was led entirely by three undergrads: myself, <a href="/phuguo/">Phillip Guo</a>, &amp; <a href="/notRichardRen/">Richard Ren</a> 
🧵1/N
Gary Marcus (@garymarcus) 's Twitter Profile Photo

It is important that we, collectively, hold OpenAI to this. The company is a nonprofit that has promised to serve humanity, and therefore it should represent humanity.

It is important that we, collectively, hold OpenAI to this.

The company is a nonprofit that has promised to serve humanity, and therefore it should represent humanity.
Abram Jackson (@abrakjamson) 's Twitter Profile Photo

thebes You laugh, but I bet someone has research in progress that will show improved performance after the model thinks you've tipped it

You are heroes (@1hopecomingtrue) 's Twitter Profile Photo

OpenAI CEO Sam Altman went on an 18-month, $85 million real-estate shopping spree — including a previously unknown Hawaii estate businessinsider.com/openai-ceo-sam…

Gary Marcus (@garymarcus) 's Twitter Profile Photo

There may be an 𝘪𝘮𝘮𝘦𝘯𝘴𝘦 clue about what led to the great OpenAI drama here. Passage from Charles Duhigg The New Yorker. If the The New Yorker ran it, it’s been fact-checked.

thebes (@voooooogel) 's Twitter Profile Photo

so a couple days ago i made a shitpost about tipping chatgpt, and someone replied "huh would this actually help performance" so i decided to test it and IT ACTUALLY WORKS WTF

so a couple days ago i made a shitpost about tipping chatgpt, and someone replied "huh would this actually help performance"

so i decided to test it and IT ACTUALLY WORKS WTF