Yeah, johnswentworth should consider an electric toothbrush with a pressure sensor. There are multiple models available for <$50.
I assumed that if you are flipping a coin, trials would be independent events and each flip would have a fixed rule (which is what happens when you flip a single coin). Instead, the coin had a different rule for odd and even numbered flips. I think that the language of the website should be amended to reflect this.
I think a useful heuristic for updating beliefs is to ask yourself "What would make this belief false?" rather than casting the issue in the framework of confirmation vs. balance. To make this concrete, consider the example of flat earthers vs. scientists. If you believed in a flat earth, there are any number of tests you could do to (e.g. watching ships sail down below the horizon) that would lead you to update towards falsifying your belief. This type of information seeking is neutral with respect to confirming your beliefs. This also allows us to look for more direct evidence around our beliefs rather than appealing to indirect methods such as whether or not a person agrees with us (see hug the query).
Second, I haven't looked into the work of Weijie Zhong, but I was wondering if there might be a bias variance tradeoff at play here for efficient information seeking (i.e. obtaining only confirmatory evidence seems likely lead to low variance but high bias)?
I'm only a vegetarian but I try to eat vegan when possible. I personally think that fortified foods are too often overlooked by people going on plant based diets. A lot of products targeted at vegans are fortified really well with nutrients that vegans tends to lack. For instance, at least at the grocery store I go to, the plant based milks contain as much calcium, b12, and vitamin D as the cow's milk. Breakfast cereals are also great and usually have an assortment of nutrients including iron (this comes with the usual caveat of avoiding the really sugary cereals). Personally, I like Cheerios and Total.
Other than that, I think it is... (read more)
Could you integrate Blackbelt with Anki or another spaced repetition framework? Someone made this set of anki cards on AI alignment sometime ago, which I have found to be a useful resource: https://www.ai-alignment-flashcards.com/decks
I often make mistakes when solving math problems without due care, whether by skipping too many steps, moving to quickly, or failing to check my work. I am aware of the weaknesses in the human mind when solving logical problems and take steps to mitigate them. By the very nature of math problems, I know when I get a wrong answer. Even if the value of a calculation is unknown, I can still check each logical step and know when I have erred. Yet when reasoning verbally, I don't get this feedback. Even if I have made several complicated steps in an argument, I may feel confident in my conclusions and most... (read more)
My tentative viewpoint on this is that the preference one has over value drift from AI vs. human made value drift comes from an entity's ability to experience joy and suffering.
In the context of AI safety, many humans who experience mostly positive lives could be killed or made to suffer at the hands of superintelligent AI in the future, and the experience of the AI in terms of pain and suffering is mostly unknown. I'm worried that an AI will optimize for something that does not lead to any subjective increase of wellbeing in the universe at the cost of human happiness.
This post mostly doesn't resonate with me, and I'm coming from a place where I really value the advice put forth in the excellent blogpost "To Listen Well, Get Curious" by Ben Kuhn. It was frustrating for me to watch "It's Not About the Nail" because I couldn't help but think about how poorly this video characterizes common misunderstandings that happen during conversations where one person is helping another. Kuhn argues that the common advice to not offer help during helper conversations is bad, since it presumes that the one being helped doesn't actually want their problem solved. He explains that this advice seems good on the surface, since many people offer... (read more)