The US should be trying to foment and support AI ethicists and safetyists defecting to China
SAFETY
-
Art Will Not Die Before We Lose Interest Humans
By
–
Not gonna happen. We are not just interested in things looking as pretty as you can make them, but in the output of your consciousness and heart. Art is not going to die before we lose interest in other human beings.
-
AGI Development for Humanity Aligns with Article Rules
By
–
I am pretty sure building AGI for the benefit of humanity fits into this specific rule written in the very article you linked
-
Grok’s Misuse Risk: xAI Must Address User Safety Concerns
By
–
[don't worry, grok is not sophisticated enough yet, otherwise I would not have posted this, but Grok's potential misuse as a tool for getting twitter users in trouble is a problem that @xAI needs to think about]
-
OpenAI Q* Breakthrough Leaked Amid Safety Concerns
By
–
4 long days after the ousting, OpenAI's secret AI model breakthrough called Q* (pronounced Q-Star) was leaked. We also found out that ahead of Sam's firing, researchers sent the board a letter warning of a new AI discovery that could "threaten humanity".
-
Sam Altman’s Chilling November 2023 Speech on AGI Nature
By
–
First, we need to go back in time to November 2023. A day before Sam was fired, he gave this chilling speech: "Is this a tool we've built or a creature we have built?" "This is the biggest update we'll have" https://
x.com/ritageleta/sta
tus/1725799427833765978/video/1
… -
Societal Impact of Open Foundation Models Position Paper
By
–
7/ On the Societal Impact of Open Foundation Models – a position paper with a focus on open foundation models and their impact, benefits, and risks.
-
Bostrom’s Information Hazards Framework for AI Safety
By
–
The paper: https://
nickbostrom.com/information-ha
zards.pdf
… -
Lebowski Theorem: Superintelligent AI Reward Function Hacking
By
–
Lebowski Theorem: "No superintelligent AI is going to bother with a task that is harder than hacking its reward function." #AGI #AGIFirst
-
Neuropsychological Hazards: Information-Based Harm and AI Safety
By
–
And then there is the science fiction favorite – the Neuropsychological Hazard, where information will actually cause physical harm. Triggering epileptics is one real example, mentioning MacBeth at a theater a fictional one. A list of more Basilisks: https://
artandpopularculture.com/Motif_of_harmf
ul_sensation
… 7/