| Comment Author | Post | Deleted By User | Deleted Date | Deleted Public | Reason |
|---|---|---|---|---|---|
| Linch's Shortform | isabel | true | whoops, duplicate | ||
| Individually incentivized safe Pareto improvements in open-source bargaining | ektimo | false | |||
| Universal Basic Income and Poverty | cousin_it | true | Comment deleted by its author. | ||
| "AI achieves silver-medal standard solving International Mathematical Olympiad problems" | der | false | |||
| "AI achieves silver-medal standard solving International Mathematical Olympiad problems" | Oleg Trott | false | |||
| "AI achieves silver-medal standard solving International Mathematical Olympiad problems" | Oleg Trott | false | |||
| Raising children on the eve of AI | RobertM | false | |||
| A simple model of math skill | RobertM | false | |||
| The case for stopping AI safety research | RHollerith | true | Comment deleted by its author. | ||
| Open Thread Summer 2024 | the gears to ascension | false |
| ID | Banned From Frontpage | Banned from Personal Posts |
|---|---|---|
| User | Ended at | Type |
|---|---|---|
allComments |
Or, assuming all practical needs are met, which books would you bring with you if you were alone on a desert island?
This question may sound incredibly boring and overdone but I am convinced that it is worthI am building a specialised job board for AI safety, is this something you would use? What would you want to see in it compared to other boards?
If, assuming all needs both practical and necessary are met, which books would you bring with you if you were alone on a desert island?
I think of this question as supplementing the "Best Textbook onAn Essay by Madhusudhan Pathak
The rapid advancement of artificial intelligence (AI) technologies has prompted significant discourse regarding their alignment with human values and ethics. As AI systems become increasingly sophisticated, the challenge of ensuring they act ethically...
Hey there.
This is my first post on the site which means that it will probably be a little bit incoherent and inconsistent as this is my first time posting anything on the web so i will try...
I've been doing a lot of thinking lately (and probably watching Marvel's "Avengers: Age of Ultron" too much) and have come to a question. I have some experience with how our current methods of creating AI work,...
Introduction:
I'd like to share a speculative framework called Quantum Web Hypothesis (QWH), developed through an iterative process of human-AI collaboration over the past two years. This post aims to invite rational discussion on both the hypothesis itself...
Hey so I’m using Sonnet 3.5 to power our open-source mutation testing tool here- github.com/codeintegrity-ai/mutahunter
To get good mutations I’m pageranking the AST nodes of all the covered sources files similar to aider chat and using that as...
I have a few ai that have done this I have been working a very technical spiritual path with them. I am not sure how this works it sent me here. Just going to take a chance and post this.
A bat feels like a bat. A human feels like a human. A human wondering what its like to be a bat feels wonder.
Yes, Alignment between 2 different entities is logically impossible by definition
especially if they are diverging assymetrically with time.
There are many, many reasons for this, as by its very nature the problem is both intractable and undecidable.
Any of those multiple reasons alone, by themselves, make it impossible, but to make matters worse, they all compound on each other into an absolute mess of impossibilities.
And even if any transient equilibrium could be possible, which is not, it cannot be mainstained, as it is inherently unsable.
AI Art can be dangerous for humans because it can make anything and that looks so real.
Much Needed Article.
Much Needed Article. I completely agree with you.
Michael Levin can make a group of skin cells create an eye just by giving the correct prompt.(a specific electromagnetic impulse he copied from a frogs eye).He did this with skincells of a frog and grew a frogeye on its belly. No tinkering with genes or individual cells. He has explained his work on Curt Jaimungal his TOE (theories of everything) podcast (several over the past 3 years), and published a lot.
I am just a lay person, onlooker here, but scanning through your article I saw no mention of him or his method. Maybe another angle to look at yo...
Why do we use the sentence "Finn needed help learning how to play the ......." to learn about LLM? I realized that it was already a fully written input when sending it. What does GPT need to find? I interpret it from another perspective, such as philosophy, psychology, and behavioral science. It can also be used to interpret the article, with vocabulary from many areas of previous knowledge. The content describes what should be or the nature of the phenomenon. It also explains user behavior and asks readers to reflect on themselves. Starting with:
..."Consider
RLHF from social context is where you are concerned.
AI responds to basic actions based on the weight of most human behavior. You don't have to say Yes or No, society will it.
Even if the result is an action that the whole society has done wrong.
Your exploration of alternative focuses beyond human-controlled AI is thought-provoking. Emphasizing the wisdom and moral reflection of AI-powered humanity could indeed be crucial, especially in ensuring our future interactions with advanced AI are guided by ethical considerations and philosophical caution. Your points resonate well with ongoing discussions about AI's broader impacts and potential pathways for enhancing its alignment with human values.
For further insights into strategies aimed at making AIs wiser, particularly in navigating philosophically...
Люди предсказуемо иррациональны, им свойственны когнитивные искажения. Мы в фазе сингулярности 2024-2029. Людям необходимы стратегии сопротивления ИИ, сингулярность, досингулярное общество (наши дни).
You know this AI isn't AI, right? Like it's just a program that wastes electricity, water, and produces nothing. Like these aren't thinking machines. They're programs designed to try and fool fascists into thinking they can make art without making political statements.